Last modified 20 months ago
Last modified on 01/17/12 12:52:37
Earthworm Release Notes V7.0 (June 15, 2006) New Modules Modifications/ Bug Fixes to Existing Modules Changes To Configuration Files And Descriptor Files Known Bugs Known Deficiencies Release Notes: Earthworm Version "working" Cleaned out after release of v6.2 on 4/15/2003 Note that v6.3 was a CVS snapshot that was released on 9/21/2005 This release is EW V7.0 and has all of the v6.3 changes and many SCNL upgrades. NEW MODULES: *********** cleandir: A contrib from Hydra for Windows. Here's the description I got from John Patton: This program is a rewrite of the old Menlo cleandir. Cleandir runs as a Windows service (as opposed to running as a Windows Scheduled Task) and monitors the directories specified in the configuration file. The criteria for deletion are as follows: If any file has a modification date/time older than the age specified in the configuration file, that file is deleted. Installed by PAF 6/06/2006 in grab_bag liss2ew_scnl: This is a contribution from HYDRA EW and is an upgrade. This has only been tested under Windows and thus only has a makefile.nt This is basically a SNCL'ized version of liss2ew but has been converted to C++. Installed by PAF 5/26/2006 ringtocoaxII: This is a contribution from HYDRA EW and is an upgrade. This has only been tested under Windows and thus only has a makefile.nt in this release. From notes that John Patton sent me: There are basically 3 differences between ring2coax and ring2coaxII: 1. ring2coaxII is multithreaded. 2. ring2coaxII has an input queue 3. ring2coaxII has my attempt at an adaptive algorithm that increases the outgoing udp data rate as the queue fills up (within configurable limits). It also tries to "smooth" the udp message load so that we're at a constant level rather then spikes and troughs (not very good at this, but better than nothing). The design idea behind it was that we had problems with ring2coaxII getting cpu starved for a few seconds and loosing data, thus the input queue. The other half was how our udp load on our network would change drastically from second to second from very high to very low, we wanted some sort of way to "smooth" it out. I've had limited success on that one. Installed by PAF 5/26/2006 pkfilter: The purpose of pkfilter is to reduce the number of "duplicate" picks from a given "station". For pkfilter, all components which have matching station code and network code are considered to be the from the same "station". The rules for what are considered "duplicate" picks can be found in the file: vX.X/src/seismic_processing/pkfilter/pkfilter.designspecs.txt Pkfilter is location-code friendly and will handle both the TYPE*2K and TYPE*SCNL versions picks and codas. LDD 5/5/2004 statrigfilter: The purpose of statrigfilter is to eliminate "duplicate" triggers from a given "station". For statrigfilter, all channels which have matching station code and network code are considered to be the from the same "station". The rules for what are considered "duplicate" triggers can be found in the file: vX.X/src/seismic_processing/statrigfilter/statrigfilter.designspecs.txt Statrigfilter is location-code friendly and will handle both TYPE_CARLSTATRIG_SCNL and TYPE_CARLSTATRIG messages. LDD 11/23/2005 putpick: Putpick is a command-line testing tool which allows one to replay a file containing picks and/or codas into an Earthworm ring. Usage: putpick Putpick reads a file containing a series of pick/coda msgs and write them to shared memory. The file can contain a mix of TYPE_PICK2K, TYPE_PICK_SCNL, TYPE_CODA2K, TYPE_CODA_SCNL, and also comments. Only valid TYPE_PICK* and TYPE_CODA* msgs will be written to shared memory; all other lines will be ignored. Putpick is useful for replaying picks into binder - binder's logfile (or a portion of it) can be used as the input file to putpick, or the picks can be placed in a separate file. If pause = y, the relative timing of writing msgs to ring is based on delta-t in the timestamps in the picks. If pause = n, the msgs are written to the ring as fast as they are read from the file. LDD 6/25/2004 ws_clientIIx: Added a new set of ws_client routines (ws_clientIIx), that are used by ora_trace_fetch. The new routines were designed to streamline and simplify access to the wave_servers while conserving wave_server resources. (The routines handle connect/disconnect for the caller, and if given a list of requests, the routines will sort the request by wave_server, so that it will connect to one wave_server, retrieve the appropriate data, disconnect from the first and connect to the second, retrieve the appropriate data, and so on. The routines were written in 2001, (derived from ws_clientII) and have never been as thoroughly tested as ws_clientII. The routines have not been scheduled for SCNLizing. DK 072004 scn2scnl,scnl2scn: A pair of conversion modules (code resides in data_exchange/scn_convert) to change trace data, picks and codas from the original Earthworm formats with site,component,network codes to the new Earthworm formats using site,component,network,location codes and visa versa. Written by Will Kohler, modified by Pete Lombard to do SCN->SCNL lookups. LDD 10/21/2004 slink2ew: A seedlink client for acquiring data from seedlink servers (e.g. IRIS) and converting to tracebuf2 (or optionally tracebuf). Much more robust than liss2ew. Written by Chad Trabant of IRIS; modified for tracebuf2 and debugged by Chad Trabant and Jim Luetgert. JHL 4/19/05 PAF Updated with version 1.3 on 1/30/2006 export_ack, export_scnl_ack: This new version of export expects to receive an acknowledgment packet for every packet it writes to the socket. Creation of this new version was desired because we observed during the export of continuous trace data that during a communication failure, export was able to successfully write several messages to the (broken?) socket, but import never received them. When the socket was re-established, those messages were lost forever. Acknowledgments are received asynchronously by the socket-reading thread. A circular buffer, SendQueue, tracks the status of msgs after they been pulled from the MessageStacker queue. If the socket is broken/reconnected, export*ack re-sends any msgs in the SendQueue which haven't been acknowledged. In normal operation, if export*ack laps the SendQueue and finds a message that has been sent but not ACKed, it will sleep 10ms and then check status again, looping until an ACK is received before preparing to send the next message. The SendQueue length is configurable, from 1 to 254 messages (default=100). Using a shorter SendQueue can cause a decrease in the rate at which export will send messages. A slower rate might be desireable so that an import system is not flooded by data after a prolonged communication break. In testing with export and import on the same machine, these are export's approximate maximum data rates: SendQueueLength = 1, 10 messages/second SendQueueLength = 10, 90 messages/second SendQueueLength = 100, 350 messages/second Using the default SendQueueLength, export*ack's data rate was not observably different than export_generic's data rate. export*ack requires using the new import_ack to communicate properly. Since the receipt of ACKs is asynchronous, we may now have the opposite problem to that we had before (duplicate data instead of lost data). It is now possible that import_ack actually received a packet, but export_ack didn't get the ACK before the socket broke. In this is case, export_ack will re-send the packet and import_ack will see the same packet more than once. The system running import_ack must be prepared for duplicate data. LDD 4/29/2005 import_ack: This new version of import is required as the data exchange partner for the new export*ack modules. However, it can communicate with either old or new-protocol export modules, and can therefore be used anywhere in place of import_generic. From the first message received per socket connection, import_ack discovers what flavor of export it's talking to. It will do either old non-ACK or new ACK protocol as appropriate on the fly. Import_ack has been modified to use same configuration commands as export for setting up the server connection and socket heartbeat strings and intervals. It still recognizes the original import_generic commands. Since the receipt of ACKs by export_ack is asynchronous, we may now have the opposite problem to that we had before (duplicate data instead of lost data). It is now possible that import_ack received a packet, but the socket broke before export_ack received the ACK. In this is case, export_ack will re-send the packet and import_ack will see the same packet more than once. Therefore, the system running import_ack as a partner to export_ack must be prepared to handle duplicate data. LDD 4/29/2005 wftimefilter: This new module was created to "sanitize" incoming waveform data streams to the liking of Earthworm modules. It reads waveform data (compressed or uncompressed) from one transport ring and writes it to another ring, filtering out duplicate data, time overlaps and bogus future timestamped data. Timestamp checks are done for each channel independently. The check for bogus future timestamps requires that the system clock is set to network time. Wftimefilter can be configured to output the messages using the original installation/moduleid or its own installation/moduleid. However, it does not alter the contents of the message in any way. Systems which are importing data using import_ack/export*ack should probably use wftimefilter to remove possible duplicate data after an import/export reconnect. Systems importing waveform data from non-Earthworm data sources should also consider running wftimefilter to ensure that only chronological timeseries enter their own Earthworm processing. LDD 5/10/2005 Modifed to allow additional optional behavior: 1)rejecting packets that have a sample-rate different than the expected one for the channel. 2)Limiting per-channel logging, to limit logfile size and increase readability. 3)Adding per-channel daily summaries of errors and warnings. 4)Adding the ability to issue a STATUS message to a ring as a part of the daily summary. Default behavior should be unchanged. DK 2006/03/27 startstop_service: (added to CVS by Paul Friberg 6/29/2005, authored by Mark Morrison) From the README.txt found in the src/system_control/startstop_sevice: startstop_service is identical to the old startstop_nt, except that it runs as a Windows service. This means that the parameters are all the same - read from startstop_nt.d - but that startstop isn't just executed from a command window or via the scheduler. Note that this version is taken from startstop_nt, so I haven't compiled or tested any of this under Solaris or other systems. I believe it's generally felt, however, that these modifications don't apply to Solaris. srparxchewsend: as a new data source. This is contributed by Symmetric Research for their digitizers. It was installed in data_sources by Paul Friberg Jan 5, 2005. rayloc_ew: (last revised 13 July, 2004) Ilya Dricker at www.isti.com Rayloc_ew is an Earthworm Module to perform global relocation of earthquakes and travel time computation. Rayloc_ew is an Earthworm-style wrapper around a Fortran program RayLoc written by Ray Bulland. Rayloc_ew attaches itself to the input ring and monitors the ring buffer for the TYPE_LOC_GLOBAL messages. Rayloc_ew reformats the input and calls FORTRAN RayLoc processing machinery. The output of rayloc_ew is TYPE_RAYLOC message, which is written to the output ring. The rayloc_ew program is compiled and tested out on a Solaris platform (with gcc and g77 compilers). startstop_unix: Added in this Linux'ified version of startstop (uses a startstop_unix.d file) from Alexandre Nercessian: PAF 9/7/2005 raypicker: Added this from the HYDRA package for Mitch Withers. This is the picker that is used by NEIC to feed glass. Only compile tested under Windows at this point since that is where NEIC uses it. PAF 01/30/2006 glass: Add this from older NEIC distro. This is a global associator written by Carl Johnson. This module will only compile/run under Windows. PAF 01/30/2006 eqfilterII: Added in this new module from Mitch WIthers for handling arc, global, and rayloc message types. PAF 01/30/2006 geqproc: Same tasks as eqproc, but this takes global messages instead of arc messages. Part of global event processing chain. PAF 01/30/2006 globalproc: Moved globalproc from NEIC v0.1 to v7.0. Recall that globalproc listens to the glass chatter, assembles quake, pick, and amp messages into a global loc message, and sends it to the ring when appropriate. Combined with geqproc, the two accomplish the same thing as what eqproc did for binder. This strays from the original intent of globalproc allowing a tweaked eqproc and globalproc to "eat" glass output simultaneously but given that we have rayloc, geqproc only needs to locate local events so this lashup was easier and does the job. Comments by Mitch Withers: PAF 01/30/2006 config_ws: This module will listen on a wave_ring for scn or scnl type trace_buf[2] messages and create a waveserverV configuration file for the tanks. It has a number of configuration options to specify how many tanks and of what size they should be. This nifty little module is contributed by John Patton of the Hydra group and installed by Paul Friberg (only compiles under Windows). PAF 02/09/2006 reconfigure: This command line tool sends a reconfigure request to startstop. Startstop will then re-read startstop_*.d and start up any new rings or modules it finds. Duplicate module names and duplicate ring names are not allowed. Necessary for this to work is the addition of Message TYPE_RECONFIG 112 to the earthworm.d file. s.lisowski 20060405 MODIFICATIONS/BUG FIXES TO EXISTING MODULES: ******************************************** hyp2000: update for Intel 9 Fortran compiler (from v6.3) note there is a makefile.nt_intel9 and a new line for linking to the intel 9 compiler in the ew_nt.cmd. For Linux, hyp2000 compiles but is still majorly broken. PAF 6/6/2006 ringdup: Added in a Hydra contribution that allows BLOCKING if particular SCNL's. This is all encapsulated in the scnfilter_exclude.c code. This compiled fine under Solaris, but has only been really tested under Windows. I did not upgrade the makefile.ux to use this yet until it has been tested. Look in the new .d file to see the notes about Block_scnl, but here's what I got from John Patton: blocker is what Hydra calls this version This module is a tweak to the program ringdup_scn. It involved adding functionality to explicitly block channels by SCNL while letting others not specified to be copied from the input ring to the output ring. The functionality is not implemented in ringdup.c itself, but in the library scnfilter. To avoid confusion, since scnfilter is used by export, this file was copied to the ringdup directory and renamed scnfilter_exclude. This file is identical to the original scnfilter except for the required changes. The makefile for ringdup was modified to link to scnfilter_exclude instead of scnfilter. The library has also recently been enhanced so that ringdup_scn understands the message type TYPE_PICK_GLOBAL. Modifications made by PAF 6/6/2006 seisan: Seisan format export in waveman2disk, trig2disk (and therefore archman, etc.) Edited seiutils.c to truncate insignificant digits off the end of a numerical comparison so as to eliminate problems where the Seisan data file would null out after a certain point in the middle of the file, and render half the file useless. startstop: Reconfigure or 'recon' command added to startstop module. The command can either be given by running the command-line 'reconfigure' tool, or by typing 'recon' or 'reconfigure' into the interactive EARTHWORM STATUS interface. Startstop will then re-read startstop_*.d and start up any new rings or modules it finds. Duplicate module names and duplicate ring names are not allowed. Necessary for this to work is the addition of Message TYPE_RECONFIG 112 to the earthworm.d file. Also some startstop commonalities for all four platforms (Windows, Windows Service, Linux, Solaris) were added to libsrc/util/startstop_lib.c, commonalities between Solaris and Linux were added to libsrc/util/startstop_unix_generic, commonalities between Windows and Windows Service were added to libsrc/winnt/startstop_winlib.c. s.lisowski 20060405 Incorporated Hydra StartstopConsole into the Windows startstop_service, and created Hydra-independent makefiles for necessary objects s.lisowski 20060606 statmgr: Modified to allow up to 10 pagegroup commands in statmgr config file. Modified descriptor files to allow optional module-specific settings for pagegroup (up to 10) and mail (up to 10) recipients. Any pagegroup or mail setting in a descriptor file overrides the statmgr config settings. This will allow the user to fine-tune on a per-module basis the recipients of various Earthworm notifications. Modified logfile name to use the name of the statmgr config file (had been hard-coded to 'statmgr*'. Modified logging of configuration and descriptor files. LDD 2006/4/25 Added new command "From" to allow user to set the "From" field of outgoing email messages. This command is used only in the Windows version of statmgr. If the "From" command is ommitted, the email "From" field will be filled using environment variables %USERNAME%@%COMPUTERNAME%. Previous versions of SendMail() defaulted to root@mailserver. LDD 2006/05/19 logit: Made it so that logit now appends a platform-specific slash to EW_LOG if there isn't one. This allows for a more consistent entry on in the environment/ew_.{cmd,bash} file. s.lisowski 20060405 gmew: Added in SCNL capability (no longer available for just SCN) and upgraded to the new version of ShakeMapXML (v3.1). I also patched a minor wave server warning message to be more explicit (empty menu) and added versioning to the program itself. PAF 03/15/2006 heli_ewII: Added in SCNL capability (no longer available for just SCN). PAF 03/15/2006 libsrc/solaris/sleep_ew.c: Added: #include WMK 3/22/04 libsrc/solaris/transport.c: Added: #include Function tport_bufthr() now returns NULL. WMK 3/22/04 libsrc/winnt/transport.c: Added: #include libsrc/util/kom.c Changed: "static struct k_buf" to "struct k_buf" WMK 3/22/04 libsrc/util/trheadconv.c New function to convert TYPE_TRACEBUF headers to TYPE_TRACEBUF2 headers. WMK 4/29/04 libsrc/winnt/sendmail.c Increased BUFFSIZE from 200 to 250, to accomodate more email recipients. This is a kludgy fix. It might better to allocate the buffer on the fly, to accommodate any number of email messages. Also, note that I did not increase BUFFSIZE in libsrc/solaris/sendmail.c WMK 6/11/03 libsrc/util/getutil.c Added new functions: GetKeyName, GetInstName, GetModIdName, GetTypeName. These functions take a numeric value (shared memory key, installation id, module id or message type) as their argument and return a pointer to its corresponding character string name from the earthworm*d tables. If the numeric value is not in the earthworm*d tables, the functions return a NULL pointer. Added new function, GetLocalInstName, which return a pointer to the environment variable EW_INSTALLATION. LDD 7/29/2004 libsrc/util/rw_strongmotionII.c: Changed rd_strongmotionII() to interpret both "-" and "?" as NULL strings when reading the QID: line qid and qauthor fields. Previously only accepted "?" as a NULL string. Change was required by CISN data which uses "-" as a NULL string in those fields. LDD 2/13/2004 adsend: Modified to produce tracebuf messages with trace2 headers. These headers contain location code and header version number. Old-style tracebuf messages (w/o location code) are no longer made. WMK 4/22/04 pick_ew: Modified to create picks and codas containing SCNLs. Input waveform messages may be of either TYPE_TRACEBUF or TYPE TRACEBUF2. WMK 4/29/04 Added new optional config file command so that pick_ew can choose which messages to process: GetLogo There's no limit to the number of GetLogo commands that can be used. If no GetLogo commands are used, then pick_ew will process all TYPE_TRACEBUF and TYPE_TRACEBUF2 messages in InRing. LDD 4/8/2005 carlstatrig: Modified for location code. Reads TYPE_TRACEBUF2 and produces TYPE_CARLSTATRIG_SCNL messages. PNL 5/5/2004 Added new optional config command so that carlstatrig can choose which messages to process: GetWavesFrom There's no limit to the nubmer of GetWavesFrom commands that can be used. If no GetWavesFrom commands are listed, then carlstatrig defaults to "GetWavesFrom INST_WILDCARD MOD_WILDCARD". LDD 4/12/2005 binder_ew: Modified to use TYPE_PICK_SCNL as input. Required changes to the station file reading library, site.c, and to the hypoinverse site file. LDD 5/17/2004 Modified to attempt pick association with entire quake list before attempting a new stack. Previously only attempted association with the 10 most recent events. This should allow picks to arrive fairly late (out-of-real-time) and still be associated with an event. This will prevent such picks from restacking into a new, duplicate event. This modification requires that binder keep track of the earliest pick sequence number associated with each quake so that no hypocenter updates are attempted if some of the supporting picks are no longer in the pick FIFO. If a Pick associates with an event that can't be updated, a TYPE_LINK message will go out, but no relocation will be done, and no new TYPE_QUAKE2K message will be issued. LDD 10/21/2004 grid.c changes: Added sanity check on mStack value; modified grid_stack to go thru pick list from newest to oldest pick. LDD 10/29/2004 bind.c: Modified to make the pick and quake FIFO lengths configurable. Added two optional config file commands: pick_fifo_length xxxx (default = 1000 = previous hardcoded length) quake_fifo_length yyy (default = 100 = previous hardcoded length) Only large networks should need to consider using these commands to increase FIFO sizes. Care must be taken to ensure that eqproc/eqprelim FIFO lengths are the same length (or larger) than binder_ew's FIFOs. LDD 3/9/2005 Modified to shut down cleanly on termination request. LDD 4/15/2005 binder_ew.c: Added two optional commands: BufferRing ringname (default = BINDER_RING = previous hardcoded string) This defines which ring binder will use for private storage space for incoming picks. ringname must exist in an earthworm.d "Ring" line. EventIdFile filename (default = quake_id.d = previous hardcoded string) This defines the name of the file that will contain the next valid eventid for binder to apply to a new association. Having these two items hardcoded had made it extremely tricky to run multiple instances of binder on the same host without them competing for common resources. Now it will be a simple matter of different configuration files to run multiple instances. This could be useful in testing different binder configurations. LDD 8/1/2005 eqproc: Modified to use TYPE_PICK_SCNL and TYPE_CODA_SCNL as input. Modified to produce TYPE_EVENT_SCNL as output. LDD 5/17/2004 Modified to make four more operational parameters configurable. All are set with new optional config file commands and default to the previous hardcoded values if their commands are ommitted. 1. pick FIFO length ('pick_fifo_length' command, default = 1000) 2. quake FIFO length ('quake_fifo_length' command, default = 100) For proper event reporting, eqproc's pick and quake FIFO lengths must greater than or equal in length to binder's FIFOs. 3. heartbeat interval ('HeartbeatInt' command, default = 0.3*tReport) 4. time interval at which quake FIFO is checked for events that are ready to send downstream ('rpt_check' command, default = 0.3*tReport) Items 3 & 4 were previously a hardcoded fraction of tReport, which is configured with the optional 'rpt_dwell' command (default = 30 s). These changes should allow eqproc to send events downstream in a more timely fashion. LDD 3/10/2005 eqprelim: Modified to use TYPE_PICK_SCNL and TYPE_CODA_SCNL as input. Modified to produce TYPE_EVENT_SCNL as output. LDD 5/17/2004 Modified to make two more operational parameters configurable. Both are set with new optional config file commands and default to the previous hardcoded values if their commands are ommitted. 1. pick FIFO length ('pick_fifo_length' command, default = 1000) 2. quake FIFO length ('quake_fifo_length' command, default = 100) For proper event reporting, eqprelim's pick and quake FIFO lengths must greater than or equal in length to binder's FIFOs. LDD 3/11/2005 eqcoda: Modified to use TYPE_EVENT_SCNL as input and to write location codes in the TYPE_HYP2000ARC message it puts out. Added new optional command "LogArcMsg" to control whether output TYPE_HYP200ARC msg is written to eqcoda's log file (0=don't log; non-zero=do log it). LDD 5/20/2004 hyp2000: Grabbed Fred Klein's most recent version of hyp2000 which is location-code compliant. LDD 7/01/2004 Grabbed Fred Klein's most recent version of hyp2000 which has a new command 'DI1' to set a different distance weighting scheme for early iterations. Normal distance weighting is still set with the 'DIS' command. This change was prompted by trouble locating the Parkfield mainshock of 9/2004. LDD 12/8/2004 Grabbed Fred Klein's most recent version of hyp2000 which has added logic to allow location codes " " and "--" to match (they both represent a "blank" location code). LDD 10/5/2005 ew2file: Added optional command "TimeBasedFilename X" to control whether a time string derived from the message contents is added to the beginning of the output file names. If X is zero (default), original file naming convention is used (SequenceNumber.suffix). If X is non-zero, a time string is prepended to the filename (YYYYMMDD-HHMMSS.SequenceNumber.suffix). This option is only applicable for TYPE_PICK2K, TYPE_PICK_SCNL, and TYPE_HYP2000ARC messages. All other message types will be named according to the original convention regardless of the TimeBaseFilename setting. Other messages types containing timestamps could be added to the new naming scheme in the future. LDD 11/11/2005 evanstrig: Modified to use TYPE_TRACEBUF2 as input and to write TYPE_LPTRIG_SCNL as output. Both message types include location codes. Changed config file command "MaxSCN" to "MaxSCNL". Added location code as the fourth argument to the "TriggerOn" command. LDD 5/21/2004 evansassoc: Added location code; modified to use TYPE_LPTRIG_SCNL as input and to write TYPE_TRIGLIST_SCNL as output. Added location code as the fourth argument to the "AddChannel" command. LDD 5/24/2004 putter: Modified to read files of TYPE_PICK_SCNL and TYPE_CODA_SCNL. Added optional second argument, ringname (name of ring to write msgs to). Previously was hardcoded to write to PICK_RING. Will still work with TYPE_PICK2K and TYPE_CODA2K msgs as long as the installation id and moduleid are less than 100. LDD 5/17/2004 findwave: 1. Modified to read both TYPE_TRACEBUF2 and TYPE_TRACEBUF msgs. 2. Modified "d" output (debug) to include pin number and message logo as both character strings from earthworm*.d and numeric values. If a field of the logo is not in the local earthworm*d files, the debug output will show one or more of these strings: inst_unlisted, mod_unlisted, type_unlisted. If the channel's installation id is different from the local installation id (set by the EW_INSTALLATION environment variable), the module_id string will read: mod_not_local. 3. Modified "w" output (wave_serverV) to include the actual installation and module id strings from earthworm*d files. If the installation is not in the earthworm_global.d file, it will be listed as INST_WILDCARD. If the moduleid is not in the earthworm.d file, it will be listed as MOD_WILDCARD. If the channel's installation id is different from the local installation id (set by the EW_INSTALLATION environment variable), the module will be listed as MOD_WILDCARD. 4. Modified all output to be space-delimited rather than tab-delimited. LDD 7/29/2004 findwave: Merged Hydra changes into findwave. The default findwave behavior is unchanged. Hydra changes include: Option: have tanks segmented by HD based on a HardDrive size Option: have findwave truncate tanksizes to 1GB, to comply with wave_serverV maximum safe tank size. Option: have findwave massage the record sizes to be a fixed size for all tanks. Option: have findwave massage the record sizes for variable-record-length channels to be slightly larger than the largest record received for that channel. Run findwave with no arguments for more information. gaplist: Added a "Total Gap" column to the output tables. "Total Gap" is the total, in seconds, of the length of all gaps for a particular channel. "Total Gap" gets reset to 0 when the day rolls over. Also, added a column named "Dead Time", which is the elapsed time since a channel died. If "Dead Time" is blank, the channel is currently alive. "Dead Time" does NOT reset to 0 when the day rolls over. WMK 1/8/04 gaplist: Slight table reformat. WMK 1/20/04 gaplist: Now accepts TYPE_TRACEBUF and/or TYPE_TRACEBUF2 messages. The config file (gaplist.d) now contains a list of SCNLs to monitor, rather than SCNs. WMK 4/30/04 gaplist: Added system-clock timestamp to logging of gaps/overlaps. Made slight formatting changes in logging. LDD 5/7/2004 gaplist: Now, the scnl array is allocated using realloc, so any number of Scnl and Label lines are permitted in the config file. Fixed two bugs: Label strings longer than 31 characters weren't null-terminated in the program. Now, they are, so long labels are correctly truncated. Also, the label array, defined in gaplist.h, had a length of 31 characters, which caused some values in the scnl array to get clobbered. This resulted in continuous dead/alive messages being sent to statmgr. WMK 5/11/04 gaplist: Added new optional command: GetLogo If no GetLogo cmds are given, gaplist will process all TYPE_TRACEBUF and TYPE_TRACEBUF2 messages (previous default behavior). Modified to allow processing of TYPE_TRACE_COMP_UA and TYPE_TRACE2_COMP_UA messages (headers are NOT compressed!). LDD 4/11/2005 evtK2ora: Fixed endTime bug in makeEwSnippet(). Thanks Jim L. Alex 5/9/03 shakemapfeed: The last two arguments to ewdb_api_GetSMDataWithChannelInfo were reversed. That meant that the number of returned SM messages would be incorrect IF that number exceeded MaxDataPerEvent. The result would be an out-of-bounds memory access and corrupted SM data. PNL 6/9/03 shakemapfeed: Changed the sorting used in Filter_SM_data. Now SM messages are sorted by SCNL instead of by idchan. That means that the output XML will have components of one station listed consecutively, a big help for shakemap. PNL 10/30/03 shakemapfeed: Changed the length of the long station name in the mapping file from 20 to 50 characters. LDD 2/5/04 skakemapfeed: (EWDB API - Strong Motion) ewdb_internal_GetAllSMMessages() (Note: This function is called by ewdb_api_GetSMDataWithChannelInfo()) Fixed a bug in PostGetAllSMMessages(). The idEvent field may be null for some records, and Oracle does not appear to be writing a 0 value to the memory area for the null field, so that record assumes the idEvent of the previous record located at that point in the internal buffer. The result is data corruption, such that you get messages that are unassociated with an Event, but they appear to be associated with an event. DK 02/12/04 shakemapfeed: Added new optional command, SMQueryMethod, to control how SM data is requested from the DBMS. Default behavior is original 2-stage query (get all SM data associated with this eventid, then get all UNassociated SM data in a space/time box). If SMQueryMethod is non-zero, shakemapfeed will request all SM data in a space/time box. This will return SM data associated with this event, SM data associated with other events, and UNassociated SM data. Regardless of query method, post-query filtering will be performed on eventid, external eventid, and external author. SM data associated with a different eventid but with the same author will be rejected. Data with a different author will be allowed - this is most likely data imported from another network and its eventids are meaningless in our DBMS. LDD 02/13/2004 shakemapfeed: Changed the auto-feed scheduling logic to allow for variable time intervals between feeds to shakemap. Feeds are scheduled at fixed times after the arrival of the archive message by using the new command, ScheduleFeed (one time per ScheduleFeed command, up to 20 allowed). Auto-feeds will continue at the last ScheduleFeed interval until UpdateDuration minutes after the event origin time. ScheduleFeed replaces the original DelayFirstFeed and UpdateInterval commands. These original commands are still accepted and will set up the auto-feed schedule in the original manner in the absence of ScheduleFeed commands. LDD 3/4/2004 k2ew: The channel bitmap bug reported below has been fixed. K2ew now reads the correct bitmap to determine which channels are streaming. PNL 6/9/03 k2ew: Version 2.38: K2ew will now, optionally, obtain network code from the K2 headers, rather than from the configuration file. For this to work, network codes need to be entered into the K2 using Altus File Assistant. A different network code may be entered for each stream. The K2 must be running application firmware version 3.02 or later. The "Network" parameter, in the config file, is now optional. WMK 3/25/05 liss2ew, ew2liss, dumpseed: There was a bug on Microsoft based systems that caused the user defined types BYTE, WORD, and LONG to be ignored. This was a problem specifically with the type WORD, which was changed from a signed 16-bit value to an unsigned 16-bit value, causing many problems, most noteably with the signed seed varibles for sample rate and sample rate multiplier. All liss based modules have had this problem corrected. JMP 06/17/2003 getfile_ew: There was a feature in getfile_ew that could cause more cpu to be used than necessary by checking time since last heartbeat more often than necessary. Added sleep_ew(1000) within heartbeat loop. MMW 06/29/2003 getfileII: Ported getfileII to FreeBSD. Two new files are in the source directory: makefile.bsd and socket_bsd.c. No other changes to files. WMK 11/25/03 getfileII: Added more string-length checking to config.c and to the incoming file names and output paths to avoid overflowing character arrays. Changed version number to 1.1. LDD 7/28/2004 sudsputaway: Added functions to properly initialize SUDS structures to Banfill's default values. CJB 10/3/2003 sudshead.h: Updated to make more compatible with suds.h in current release of PCSUDS. CJB 10/3/2003 file2ew: Fixed bug in raw2ew.c which caused file2ew to exit when the SuffixType command contained only 2 arguments. The third argument (installation ID) was supposed to be optional, and now is! LDD 10/3/2003 ew2file: Fixed bug in timing of heartbeat file generation. Had been using HeartBeatInt instead of HeartFileInt to control the production loop. Caused continuous heartbeat file generation when HeartFileInt==0. LDD 11/24/2003 sniffwave: Changed to sniff headers of both TYPE_TRACEBUF and TYPE_TRACE_COMP_UA msgs. Will only print data values for TYPE_TRACEBUF messages. Changed to accept the any of these strings as wildcards in SCN: "wild", "WILD", "*". LDD 10/31/2003 Modified to list both the Quality value and (if TRACE2_HEADER the Version value) in the header output. DK 2005/03/27 sniffwave: Modified to sniff headers of TYPE_TRACEBUF2 and TYPE_TRACE2_COMP_UA msgs. Added new command-line argument to support these new message types. LDD 4/13/2004 vX.X/src/data_exchange/Makefile: Added directives for STANDALONE_MODULES (right now getfileII and sendfileII) so that their executables, which are created in their source directories, are also copied to the vX.X/bin directory. LDD 11/27/2003 sendfileII: Ported to BSD Unix. Added a new makefile, named makefile.bsd. To compile, type "make -f makefile.bsd". Otherwise, everything is the same as on Solaris. WMK 2/17/04. sendfileII: Added optional config command "-SendPause xxx" where xxx is a time in milliseconds (default = 0). sendfileII will sleep SendPause millseconds before trying to open a file for shipping. This gives file writers of unknown behavior some time to finish writing and to close the file before sendfileII wisks it away. Windows fopen_excl() was supposed to keep us from opening a file that another process was using, but I don't think it's working the way we expected. Besides, Solaris never had the fopen_excl() option, so SendPause helps us there. LDD 3/23/2006 EWDB - Station Information: Modified the two stored procedures that handle the processing of SCNL from station list tools (stalist_XXX2ora) and automatic data (trace feed, automatic pick feed, strong motion, etc.). Modified the procedures to properly process the Location Code of the SCNL. Prior to this, Location codes weren't properly being handled. Because this change is to SQL procedures and not C code, you must refresh the DB stored procedures by updating with the SQL scripts from "working". (run ewdb_load_external_utils.sql from schema/sql_scripts). DK 02/17/04 naqs2ew: Modified to produce TYPE_TRACEBUF2 messages to support location codes. Can optionally be configured to produce the original TYPE_TRACEBUF msg. New config file commands: OutputMsgType (optional) and RequestChanSCNL. See sample configuration file for details. LDD 4/14/2004 Modified to output all packets (regardless of time-order) to Earthworm when RequestChanSCNL "delay" is -1. This will allow out-of-chronological order data and possibly duplicate data into the EW system. For delay>=0, out-of-order packets will be ignored (previous behavior). LDD 11/22/2005 Added new command "RepackageNmx" (optional) to control how Nmx packets are repackaged into Earthworm messages. If set to 1, naqs2ew will create one EW msg from each Nmx packet. If the Nmx packet contains more than MaxSamplePerMsg samples, multiple EW msgs will be created. The entire Nmx packet will be converted immediately. If set to 2 (the default), naqs2ew will repackage Nmx packets into 1-second EW msgs. If an Nmx packet contains more or less than 1-second of data, some samples will be buffered until the receipt of the next Nmx packet for that channel. Some shorter EW messages will be generated if time tears are present. LDD 11/23/2005 Fixed bug in how compression bits of raw Nmx compressed data packets are interpreted in unpack_tsdata_bundle() of nmxp_packet.c. Now, if both compression bits are 0, it is interpreted as "this data set contains no data" and the 4-byte data set is ignored. Previously, if both compression bits were 0, it was (mis)intrepreted to mean that compression was not used and that the data set contained a single non-differenced data value. This bug would have added extra bogus samples to the output, but probably only when a data packet was partially full. Since I can't think of a reason why a datalogger would not fill each data packet fully, I'm guessing that this bug was very rarely, if ever, encountered. LDD 1/4/2006 reftek2ew: Modified to produce TYPE_TRACEBUF2 messages to support location codes. Config command "SCNFile" was replaced by the "SCNLFile" command. Location code was added to the SCNLFile (which maps between DAS#,stream,channel and SCNL,pin) as the 7th of 8 fields on each line. Added a new config file command "SendUnknownChan" to control whether channels not listed in the SCNLFile are written as TYPE_TRACEBUF2 msgs. If SendUnknownChan is non-zero, reftek2ew makes up the SCNL as follows: Station is the DAS serial number, component is "Cxx" where xx is the channel number from the DAS, network is "Sy" where y is the stream number from the DAS, and location code is "--" (the NULL location). If SendUnknownChan is zero, data from channels not listed in SCNLFile is ignored. LDD 5/25/2004 compress_UA: Modified to produce TYPE_TRACE2_COMP_UA messages when handling TYPE_TRACEBUF2 messages as input. Compress_UA can still take TYPE_TRACEBUF packets as input and will produce TYPE_TRACE_COMP_UA messages as output. Eliminated "GetWavesFrom" command and replaced it with two new commands: "GetLogo" to supply the logos to process and "CompressSCNL" to supply the list of channels to process. If handling TYPE_TRACEBUF messages, the location code in the CompressSCNL command is ignored, but it still must be provided. LDD 4/16/2004 decompress_UA: Modified to work on both TYPE_TRACE2_COMP_UA and TYPE_TRACE_COMP_UA msgs. Produces TYPE_TRACEBUF2 messages if input is TYPE_TRACE2_COMP_UA. Produces TYPE_TRACEBUF messages is input is TYPE_TRACE_COMP_UA. Eliminated "GetWavesFrom" command and replaced it with "GetLogo" command. LDD 4/16/2004 decimate: Modified to work on either TYPE_TRACEBUF and/or TYPE_TRACEBUF2 messages to support the addition of location codes to network processing. Output message will be of same type as input message. Changed the GetWavesFrom command to require a third argument specifying the input message type (must be either TYPE_TRACEBUF or TYPE_TRACEBUF2). If the input message type is TYPE_TRACEBUF, location codes are ignored. Replaced the GetSCN command with GetSCNL to support the addition of location codes. If the GetSCN command is used, the input and output location codes are set to "--". LDD 5/11/2004 sniffring: Modified so it does not attempt to print the contents of two new binary message types: TYPE_TRACEBUF2 and TYPE_TRACE2_COMP_UA. LDD 4/16/2004 export_actv, export_scn_actv: Moved source code from data_exchange/export_actv into data_exchange/export to avoid duplication of the supporting export filtering code. The versions of the filters in export_actv had not had bug fixes in quite a while and were not location-code compliant. LDD 4/20/2004 export_scn, export_scn_pri, export_scn_actv: Modified the scnfilter code to include location code as well. Replaced Send_scn and Send_scn_remap config commands with Send_scnl and Send_scnl_remap. If old commands are used, the location code will be set to * (wildcard). Modified to work with all the new location-code friendly Earthworm messages: TYPE_TRACEBUF2, TYPE_TRACE2_COMP_UA, TYPE_PICK_SCNL and TYPE_CODA_SCNL. Still works with old format messages. LDD 4/21/2004 export_generic,export_scn Added logging under "Verbose" mode concerning the queue status of each message to be exported. This will generate VERY LARGE log files, but could be useful in debugging. LDD 3/23/2005 write_cta: Program now reads TYPE_TRACEBUF and TYPE_TRACEBUF2 messages from the transport ring. All TYPE_TRACEBUF messages are converted to TYPE_TRACEBUF2, before writing to tape. Tape files now begin with "wu" (instead of "wt"), to indicate that they contain TYPE_TRACEBUF2 messages. WMK 5/11/04 read_cta: Program now reads tape files whose names begin with "wt" and "wu". "wu" files contain TYPE_TRACEBUF2 messages. WMK 5/11/04 cta2cusp: Cosmetic changes. Program will process cta files beginning with wt or wu. Only old version triglist messages, without location code, will work. If there are multiple channels with the same SCN, they will be interleaved in the file sent to CUSP. This is bad. WMK 5/12/04 tankplayer: Added support for TypeTraceBuf2, which was missing from the get_time function that is used to determine the time of the packet from the trace header. Changed the program's behavior when Tracebuf2 packets are encountered with VERSION fields that don't match those compiled into the program. Previously the program always quit at the first sign of a tracebuf with a mismatched version number. Now depending on the value of the IgnoreTBVersionNumbers config command, the program attempts to parse and either play or skip the message based on the value of the command. (TRUE = play, FALSE= ignore). Added a log message when tankplayer comes across a large delay between packets. Modified the sleep code that executes between packets, so that it sleeps in small intervals and continues to issue heartbeats even during a large gap. Added additional debugging info for dealing with problems encountered in tracebuf files. DK 2006/03/09 tankplayer: odified to read TYPE_TRACEBUF2 files in addition to the original file types (TYPE_TRACEBUF and TYPE_ADBUF). LDD 5/21/2004 remux_tbuf: Added WaveMsg2MakeLocal() call to report and filter out tracebufs with headers that don't checksum. Replaced fprintf(stdout) commands with logit(). Added a progress indicator that prints a ".' to the screen each time 100 tracebufs have been processed. DK 2006/03/09 time_ew.c (timegm_ew()) Fixed bugs in timegm_ew() that appeared on machines not set to UTC timezone. DK 07/12/2004 import_generic: Moved the generation of Earthworm TYPE_HEARTBEAT messages out of the Heartbeat thread and into the main() thread. EW heartbeats will be generated in both the "connect to server" loop and the main working loop. This means that import will keep beating its EW heart when its export server becomes inaccessible, avoiding unnecessary restarts by statmgr/startstop. The former Heartbeat thread was renamed the SocketHeartbeat thread because its sole job is to write heartbeats (alive strings) to the socket. Also changed names of variables related to the socket heartbeat thread to avoid confusion. LDD 3/15/2005 Restructured the "connect to server" loop to be easier to follow and reworded the messages logged during connection attempts. LDD 3/17/2005 import_gen_pasv: Incorporated Richard Godbee's bug fixes in the accept_ew loop (fixed mis-placed end-while bracket, uninitialized variables) and increased the length of the RingName and MyModName variables. Reformatted the entire file with proper, readable indenting, changed accept_ew timeout to HeartBeatInt (had been hard-coded to 9 seconds), added inter-connection time interval to logging, generally cleaned up comments. LDD 4/13/2006 Most Modules: Linux Makefiles (makefile.ux) have been added to a number of modules and in some cases _LINUX compile directives have been added into the code. The list of codes to which this has been done will be provided as a separate note. 8/11/2005 Paul Friberg startstop_nt: Added in restart command to Interactive() function. This nice modification was provided by Murray McGowan. Tested and added by Paul Friberg 9/7/2005 startstop_unix/startstop_solaris: Added in the same restart command as above. It's a nice feature to have. (9/7/2005) CHANGES TO CONFIGURATION FILES and DESCRIPTOR FILES: **************************************************** earthworm.d: Added this earthworm type to allow sending a reconfigure command through rings. Message TYPE_RECONFIG 112 statmgr.d: Up to 10 pagegroup commands are now allowed (at least one is required). LDD 2006/4/25 *.desc: Modified descriptor files to allow optional module-specific settings for pagegroup (up to 10) and mail (up to 10) recipients. Any "pagegroup" or "mail" commands in a descriptor file override the statmgr config settings. This will allow the user to fine-tune on a per-module basis the recipients of various Earthworm notifications. LDD 2006/4/25 shakemapfeed.d: Added new optional command, SMQueryMethod, to control how SM data is requested from the DBMS. Default behavior (SMQueryMethod 0) is original 2-stage query (get all SM data associated with this eventid, then get all UNassociated SM data in a space/time box). If SMQueryMethod is non-zero, shakemapfeed will request all SM data in a space/time box. This will return SM data associated with this event, SM data associated with other events, and UNassociated SM data. Regardless of query method, post-query filtering will be performed on eventid, external eventid, and external author. SM data associated with a different eventid but with the same author will be rejected. Data with a different author will be allowed - this is most likely data imported from another network and its eventids are meaningless in our DBMS. LDD 02/13/2004 shakemapfeed.d: New command "ScheduleFeed <minutes(float) after receiving event>" sets the shakemap feeding schedule. Up to 20 ScheduleFeed commands are allowed, setting one time per command. Auto-feeds will continue at the last ScheduleFeed interval until UpdateDuration minutes after the event origin time. ScheduleFeed replaces the original DelayFirstFeed and UpdateInterval commands. These original commands are still accepted and will set up the auto-feed schedule in the original manner in the absence of ScheduleFeed commands. LDD 3/4/2004 naqs2ew.d: Two new commands to support location code addition: "OutputMsgType " to choose output format. If the OutputMsgType command is omitted, naqs2ew will produce TYPE_TRACEBUF2 messages by default. "RequestChanSCNL" replaces the "RequestChannel" command. It has all the same arguments with the addition of the argument. LDD 4/14/2004 binder_ew.d: Two new optional commands: "pick_fifo_length " set the length of the pick FIFO (default=1000) "quake_fifo_length " sets the length of the quake FIFO (default=100). If a command is omitted, the corresponding fifo will be set to the previous hardcoded length. LDD 2/9/2005 eqproc.d Four new optional commands. If any of these commands is omitted, the corresponding parameter will be set to its previous hardcoded value. "pick_fifo_length " set the length of the pick FIFO (default=1000) "quake_fifo_length " sets the length of the quake FIFO (default=100). For proper event reporting, eqproc's pick/quake FIFO lengths must be greater than or equal to binder's pick/quake FIFO lengths! "HeartbeatInt " sets the interval (seconds) between TYPE_HEARTBEAT messages (default=0.3*tReport). "rpt_check" " sets the interval (seconds) at which the quake FIFO will be inspected to see if it's time to send an event to downstream modules (default=0.3*tReport). The parameter tReport is set with the 'rpt_dwell' command (default=30s). LDD 3/11/2005 eqprelim.d Added two optional commands. If either of these commands is omitted, the corresponding parameter will be set to its previous hardcoded value. "pick_fifo_length " set the length of the pick FIFO (default=1000) "quake_fifo_length " sets the length of the quake FIFO (default=100). For proper event reporting, eqprelim's pick/quake FIFO lengths must be greater than or equal to binder's pick/quake FIFO lengths! LDD 3/11/2005 compress_UA.d: Eliminated "GetWavesFrom" command and replaced it with two new commands: "GetLogo " Inst_id and/or Mod_id fields can be wildcards, but msg_type field must be either TYPE_TRACEBUF or TYPE_TRACEBUF2. Multiple GetLogo commands may be supplied (no limit on number). "CompressSCNL " The wildcard character '*' (without quotes) can be used in any of the fields. Use as many CompressSCNL commands as needed to define the list of channels to be processed. If input is TYPE_TRACEBUF, the field will be ignored, but it must still be provided. LDD 4/16/2004 decompress_UA.d: Eliminated "GetWavesFrom" command and replaced it with new command: "GetLogo " Inst_id and/or Mod_id fields can be wildcards, but msg_type field must be either TYPE_TRACE_COMP_UA or TYPE_TRACE2_COMP_UA. Multiple GetLogo commands may be supplied (no limit on number). LDD 4/16/2004 decimate.d: GetWavesFrom: added a third requires argument specifying the input message type. Only valid messages types are TYPE_TRACEBUF and TYPE_TRACEBUF2. Replaced the GetSCN command with GetSCNL to support the addition of location codes. GetSCNL requires 8 arguments, input SCNL and output SCNL. No wildcards are permitted. LDD 5/11/2004 export_scn*.d: Added two new commands, Send_scnl and Send_scnl_remap, to implement location code filtering. If old commands (Send_scn and Send_scn_remap) are used, the location code is set to * (wildcard). LDD 4/21/2004 wave_serverV: Modified behavior so that wave_server does not Abort On A Single Tank Failure, unless the AbortOnSingleTankFailure flag is set in config file. (This was the documented behavior). wave_serverV: Wave server now limits tank sizes to 1GB. This applies only to NEW tanks, not existing tankfiles. The default behavior is for wave server to EXIT if it comes across a tank in the config file that is >= 1GB. If the "TruncateTanksTo1GBAndContinue" command is set in the config file, then wave server will trunctate any >= 1GB tank files to just under 1GB, and continue normal operation. This was done to prevent users from creating unsafe large tanks, after a bug was discovered that limits the safe wave server tank size to ~1GB. DK 2005/03/17 Added code to check the return value from call to WaveMsg2MakeLocal(), and reject packets for which WaveMsg2MakeLocal() reported and error. Updated the wave_serverV version and timestamp. This will cause wave_serverV to filter out packets with header issues that weren't previously filtered out before. Fixed a bug where wave_serverV would filter out packets that contained only one sample. DK 2006/03/27 wave_serverV / wave_viewer / ws_clientII.c: Modified wave_server protocol slightly to better differentiate between SCN vs. SCNL clients and servers. New wave_server clients should be backwards compatible with previous SCNL wave_servers, but the new wave_serverV is not backwards compatible with any clients. The new wave_viewer works with ALL wave_servers (SCN, old SCNL, new SCNL), and the new ws_clientII library works with ALL SCNL wave_servers. DK 2005/04/22 wave_serverV / wave_viewer / ws_clientII.c: Fixed bug in wave_server (server_thread.c ParseMsg()) that was incorrectly reading the Ascii FILL value from the "endtime" field instead of the "fill" field. This fix only affects ascii wave_server clients on gappy channels. Fixed more bugs in the ASCII protocol logic between ws_clientII and wave_serverV. DK 2005/09/07 ringtocoax: Changed tport_getmsg to tport_copyfrom so that it can distinguish missed messages (overlapped in ring) from sequence gaps in the ring. This can be useful for fine-tuning the BurstCount and BurstInterval configuration settings. LDD 2005/4/8 localmag: Does not compile under latest release of earthworm (needs to be SCNL'ized). It will compile with earlier versions however and 3 new options have been added in: useMedian, minStationsMl, require2Horizontals which are all explained in the localmag/Doc/localmag_cmd.html file. The program now has a version number to distinguish different released versions. Upgraded as per requests from Univ of Utah and HVO on 8/10/2005 by Paul Friberg (PAF) I upgraded localmag to use SCNL now and tested it at Utah 3/01/2006 PAF KNOWN BUGS: *********** liss2ew: liss2ew has been observed intermittantly producing malformed TRACE_BUF messages. Currently the conditions for causing this problem are unknown. Due to this, liss2ew should be treated as suspect. Use at your own risk. JMP 6-18-2003 k2ew: k2ew uses the k2hdr.rwParms.misc.channel_bitmap parameter to decide which channels it will see as streams from the K2. This is actually the parameter which the K2 uses to decide which channels to record in an event file (see K2's command). The K2's is the parameter that shows which channels the K2 is streaming out. This is the parameter that k2ew should be using to decide what data it will see. The big problem is that none of the header files from Kinemetrics (nkwhdrs.h) seem to show this parameter anywhere. This issue is only a problem for those folks who want to record more channels on the K2 than they want to stream continuously. Terry Dye (Univ of Utah) discovered this problem when trying to stream only 3 channels of a 12 channel K2. LDD 4/11/2002 FIXED PNL, 6/9/03 wave_serverV: it occasionally sends the following error to statmgr: UTC_Thu Sep 06 03:30:14 2001 WSV1/wave_serverV_nano ReadBlockData failed for tank [z:\nano57.tnk], with offset[999908] and record size[64]! errno[0] Mail sent. The nominal tank size is 1 megabyte, and the actual tank size is 999908. It looks like waveserver is trying to read off the end of the tank. WMK 9/6/01 wave_serverV: appears not to reply to requests for a single sample of data. I noticed when testing wave_viewer, that if the start time and end time of a request were equal (in which case there should be one sample of data) then wave_serverV did not reply to the request (ASCII request) at all. No Data, No Flags, No Reply, No Nothing. It needs to issue a reply to every request. DavidK 09/25/01 wave_serverV: Wave server appears to be limited (on 32-bit systems) to 1GB tank sizes. A bug (a signed integer overflow of tmpoE at Line 1155 of serve_trace.c (LocateRoughOffset())) prevents data retrieval from the tank under certain conditions when the tank size exceeds 1 GB. DavidK & PNL 09/27/04 Automatic restarts of adsend (using the "restartMe" line in the descriptor file) can cause an NT system to hang. Therefore, you should never use the autorestart feature with adsend, but you should bring down the entire Earthworm system if adsend needs to be restarted. LDD 5/31/2000 Comments added to adsend.desc, but leave this warning here! libsrc/utils/site.c: The strings used for station, channel and network are required to be fixed length with trailing spaces added to short names. If the strings given to site_index do not have these trailing blanks, SCN names will not match. This is not documented anywhere. PNL 10/15/00 socket_ew: (libsrc/util/socket_ew_common.c libsrc/solaris/socket_ew.c libsrc/winnt/socket_ew.c include/socket_ew.h) Fixed a bug in the connect_ew function(). When run in non-blocking mode (clients connecting to a server - using a timeout value), there was a bug in the connection code(discovered on Solaris) that caused the function to return a timeout-error when there was any kind of error during connection. The bug was discovered when connecting to a non-existent socket. When trying to connect to a non-existent wave_server on a machine, the underlying socket library was returning a Connection-Refused error, but the socket_ew library was passing back a timeout error. - The change only applies when a socket-error occurs while the function is waiting for the connect to happen. The instance where you will most-likely see a difference, is when you try connecting to a non-existent socket. Previously the function would return TIMEOUT, now it will return connection REFUSED. - Added a new function: socketSetError_ew(), and defined a new socket_ew return code: CONNREFUSED_EW. DK 2003/02/04 sm_ew2ora: There is a bug in sm_ew2ora that involves having multipe time intervals for components and channels. If a strong motion message containing information for a channel that the DB has never seen before is loaded into the DB, and then later another message for the same channel with an earlier timestamp is loaded, the load of the second message will fail, due to problems with overlapping time intervals, and the call that sm_ew2ora uses to create those time intervals. This problem only affects stations that were not previously loaded via one of the station loader programs stalist*2ora, and only when receiving SM data that is timestamped with a time that is prior to the original time for that channel. The bug lies in the logic of ewdb_api_PutSMMessage(), and not in the underlying code. Davidk 05/25/01 A change was made to ewdb_api_PutSMMessage() that dramatically affects sm_ew2ora. Please see the note about that function. Davidk 2001/07/26 ewdb_api_CreateWaveform() (ewdb_internal_CreateSnippet.c ewdb_internal_CreateWaveformDesc.c) Added a call to release the cursors used by the internal functions, when they fail. Fixes a bug which resulted in a DB cursor leak when a call to stuff a snippet into the DB failed. DK 2003/02/04 logit.c: 1) Added a new function logit_core() int logit_core( char *flag, char *format, va_list ap); This function is the same as logit(), except that it takes a va_list argument for the variable length parameter instead of '...', and it has an int return value. This function can be called by other functions that receive a '...' variable argument list, where as logit() cannot. logit_core() is to logit() what vsprintf() is to sprintf(). - 2) Moved all of the functionality in logit() to logit_core(), and modify logit to call logit_core(). - 3) Added a new function get_prog_name2(), as an intended replacement of get_prog_name(). get_prog_name2() includes an additional parameter (the buffer length of the output buffer). - 4) Modified logit_init() to use get_prog_name2() instead of get_prog_name() - All four(4) of these changes should be backwards compatible with all existing earthworm code, and the modified logit.c has already been tested with several earthworm modules. - ewdb_api: Added new function ewdb_api_GetEventSummaryInfo() to retrieve summary information for an event(including the preferred szSource and szSourceEventID of the event). 011904 DK KNOWN DEFICIENCIES: ******************* ringtocoax: In Windows NT, the time resolution of sleep_ew() is about 16 msec (one clock tick). On Solaris, the resolution is about 10 msec. This is a problem for ringtocoax, since packet delays need to be set to a few milliseconds. statmgr: A space is needed between "tsec:" and the value. If it isn't there, things fail. Artifcat of the kom routines. (Alex) threads functions: The KillThread function on WindowsNT and Solaris terminate the thread without ensuring that no mutexes are held. If a thread holds a mutex when it dies, no other thread can get that mutex. PNL 1/12/2000 carlsubtrig: The system time must be set to GMT and ew_nt.cmd must have TZ=GMT for carlsubtrig to work. Comments in ew_nt.cmd done 5/25/00. Barbara localmag: needs GMT set on the system ew2seisvole: on NT, exits with horrible crash when system is stopped. NUMBER OF RINGS LIMITED ON SOLARIS: Under Solaris 2.6 (and probably other versions as well), the maximum number of shared memory segments is six. This means that on an out-of-the-box machine you can only configure six rings. If you try to configure more than that, you will see a cryptic message from tport_create about too many open files. The fix to this problem is to add the following lines to the /etc/system file, and then reboot the system. set shmsys:shminfo_shmmax = 4294967295 set shmsys:shminfo_shmmin = 1 set shmsys:shminfo_shmmni = 100 set shmsys:shminfo_shmseg = 20 set semsys:seminfo_semmns = 200 set semsys:seminfo_semmni = 70 This allows for 20 rings. Lucky Vidmar (7/6/2000) startstop_solaris: Fixed bug in call to logit_init(). The program was printing error messages: Invalid arguments passed in. Call to get_prog_name failed. WARNING: Call logit_init before logit. startstop_solaris: There MAY be a problem with the signal that startstop sends to modules during the shutdown sequence. The shutdown sequence is started (after typing "quit" to startstop or running "pau") by startstop placing a terminate message on all transport rings. Modules should see this message and start their own shutdown. After a configurable delay, startstop checks to see that all modules have exitted. Any that are still running are sent a signal to terminate them. Currently that signal is SIG_TERM. But since wave_serverV has a handler for SIG_TERM, wave_serverV sees that as essentially the same as a terminate message. So if wave_serverV is having problems completing its shutdown, SIG_TERM won't do anything. The result is that startstop may give up and exit, leaving wave_serverV running. If that happens, the operator will have to terminate wave_serverV by doing "kill -9 ". That may leave shared memory and semaphores stranded in the kernel: run the command "ipcs -a" to see. If necessary, the stranded shared memory and semaphores may be cleaned up with the ipcrc command; must be run as root; see the man page. This problem only exists on Solaris/Unix, not on WindowsNT. PNL, 10/4/2000 libsrc/utils/kom.c: fix comment in k_open() The comment above k_open() says that only one file can be open at a time. Yet the Kbuf array has slots for MAXBUF (currently 4) open files. Does this work, or is the comment to be taken at it's word? PNL 10/15/00 libsrc/utils/logit.c: logit_init() requires a module_id number, which it uses to construct the log file name. This is not helpful, since the module_id number is not meaningful to people. Worse, it requires that the config file be read and earthworm.d lookups be completed before logit calls can be made. Thus errors in the config file can only be reported to stderr or stdout instead of being saved in a file. PNL 11/29/00 libsrc/util/k2evt2ew.c: This library supports a maximum EVT data size of 800000 samples per channel per EVT file. This value is hardcoded as MAXTRACELTH in include/k2evt2ew.h. k2evt2ew() will return a warning if the EVT file exceeds this size, and process as much as the EVT file as the hardcoded limit allows. DK 2003/01/18 import_ida: Modified to build under WIN32(#ifdef'd a couple of includes and modified the NT makefiles.) (saved under tag HYDRA_V1-20) Modified for SCNL and Tracebuf2. DK 2005/04/22 TRACEBUF messages. The definition of `endtime' of the TRACEBUF message is not documented. Some programmers are taking it as the "expected start time of the next TRACEBUF packet (if the sample interval is uniform.)" The more accepted practice is that `endtime' is the time of the last sample of the current TRACEBUF packet; that is, one sample interval less than the expected start time of the next TRACEBUF messsage. Using this last definition, if a TRACEBUF packet has exactly one sample, then its starttime and endtime are the identical. Clearly this distinction needs to be documented. The file waveform_format (in the /home/earthworm/DOC directory) gives no specifics about start or end times. PNL 1/24/01