source: trunk/release_notes.v7.3 @ 5054

Revision 3513, 9.0 KB checked in by paulf, 10 years ago (diff)

moved working to v7.3

  • Property svn:eol-style set to native
  • Property svn:keywords set to Author Date Id Revision
Line 
1Release Notes:  Earthworm Version "working"
2Cleaned out after release of v7.2 on 12/28/2007
3
4These notes are for release v7.3  FINALIZED on 12/23/2008
5
6NEW MODULES:
7***********
8SaraAdSend - new module to import data from a SARA digitizer
9over a serial port. Written by Larry Cochrane (webtronics).
10Imported into EW by Paul Friberg 2008-02-11
11
12RockToEW - Provides a data relay from a Kinemetrics-Rockhound-compatible
13instrument into Earthworm via an export of Tracebuf2 messages
14Written by Eric Thomas of ISTI 2008-11-06
15
16MODIFICATIONS/BUG FIXES TO EXISTING MODULES:
17********************************************
18raypicker - upgrade from Carol Bryan, latest version from NEIC
19Paul Friberg 2008-12-02
20
21srparxchewsend - with direction from Symmetric, downloaded latest code
22from
23http://www.symres.com/support/download.htm
24installed and and tested compiles on Windows and Linux.
25Stefan Lisowski 2008-12-02
26
27localmag - Claudio Satriano noticed a bug in the SAC input file
28reading that was causing seg faults. This is now fixed in the
29latest version.
30Paul Friberg 2008-12-01
31
32tankplayer - added a GetFromDir option to the module so that one could
33load tanks dynamically. This is a way for groups who have strange format
34data to convert them into tanks and run them in near real time. This
35code borrows heavily from the file2ew implementation. It has been tested
36in both old style (listing WaveFile entries) and with GetFromDir mode.
37The Documentation will be updated to reflect the new mode in v7.3 of ew.
38Paul Friberg 2008-09-09
39
40sniffring - fixed a weird case where a module id was foreign to a
41local installation (imported) and thus not lookupable in the earthworm.d
42module list. Verbose mode would core dump in this case.
43Paul Friberg 2008-07-31
44
45naqs2ew - updated the list of Nanometrics Instrument types to include
46all current Nmx hardware.  Lynn Dietz 2008-07-21
47
48scream2ew - Modified to issue a single Earthworm heartbeat at the end
49of ewc_init(), which is called just before a connection to the
50SCREAM server is attempted. Previously, the first Earthworm heartbeat
51was issued only after the SCREAM connection was complete. If the
52SCREAM connection failed, scream2ew exited with no heartbeat, making
53it impossible for statmgr to issue an auto-restart request because
54it didn't know scream2ew's PID. Now if the SCREAM connection fails,
55auto-restarts via statmgr/startstop will be possible.
56Lynn Dietz 2008-06-26
57
58startstop - for UNIX, if you had a bug in your startstop_sol.d or
59startstop_unix.d configuration files, then the program would stop and
60issue a bogus message about logit_init() not set. However, if you
61put the .d file on the command line, it would intelligently process
62the .d file and warn you where the problem was. I fixed this by
63modifying the startstop_unix_generic.c library call.
64Paul Friberg 2008-05-19
65
66k2ew_tcp - Lynn Dietz discovered and fixed a compilation issue for special
67cases of k2s that have modems and serial streaming (used at CGS).
68
69trig2disk - fixed starttime of additional SCNL includes. It was incorrectly
70providing the max start time of the TRIGGER message. It now correctly uses
71the minimum start time of the SCNLs in the TRIGGER message for any
72additional SCNLs. Bug spotted by Franco Mele of INGV, fixed by Paul
73Friberg 2008-04-28
74
75ms2tank - improved code so that it can handle mulitplexed mseed packets.
76Paul Friberg 2008-04-22
77
78binder_ew - added optional command "BufferRingKB" to control the size of
79binder's private buffer ring. The default is 256 KB (262144 bytes) is
80slightly larger than the previous hard-coded size (250000 bytes).
81Lynn Dietz 2008-04-22
82
83src/libsrc/util/libmseed - upgraded to 2.1.4 version of libmseed library
84from Chad Trabant of IRIS DMC.
85Paul Friberg 2008-04-04
86
87pick_ew - added a PickIndexDir optional setting to allow the pick_ew_MMM.ndx
88files to be written somewhere other than the params directory. The params
89directory is getting cluttered up enough and its nice to have the option
90to write these indexes elsewhere.
91Paul Friberg 2008-03-28
92
93slink2ew - 2.0ewd - upgrade from Chad Trabant that inclues Matteo's
94fix and some new features for wildcarding (Paul Friberg 2008-03-14)
95slink2ew - 1.4.1 - fixed a crashing problem that involved bad
96mseed packets not being detected. They are now skipped.
97Matteo Quintilani 2008-03-05
98
99waveman2disk - upgraded the logit_init() call to be able to handle
100the largest possible trigger message size MAX_TRIG_BYTES. Before
101it was set to just 1024 and this would throw out a bogus error
102message that the logit buffer was too small. Reported by
103Jean-Michel Douchain.
104Paul Friberg 2008-02-25
105
106carlstatrig - would report that it was not using STAtime in an error
107message when it actually was! This was fixed by David Wilson of HVO
108and is a minor change to the readcnfg.c module.
109Paul Friberg 2008-02-21
110
111fir - fixed a byte-order issue that appeared when the WaveMsg2MakeLocal()
112was called before some header values were checked. This resulted in
113a failure that indicated zero length traces.
114Mitch Withers 2008-02-13
115
116reftek2ew - add makefile.ux
117Richard Godbee 2008-02-11
118
119sniffwave - added in a new simple command line option to replicate:
120        sniffwave RING wild wild wild wild n
121by simply typing:
122        sniffwave RING
123This will hopefully save folks just wishing to look at the tracebufs
124in a ring sometime.
125Paul Friberg 2008-01-19
126
127putpick - didn't have a makefile.ux, now it does
128Paul Friberg 2008-01-16
129
130localmag - fixed a bug with the LM option of outputFormat. I am
131shocked this was never discovered before...basically an if should
132have been an "else if".
133Paul Friberg 2008-01-08
134
135CHANGES TO CONFIGURATION FILES and DESCRIPTOR FILES:
136****************************************************
137
138KNOWN BUGS:
139***********
140
141liss2ew:
142liss2ew has been observed intermittantly producing malformed
143TRACE_BUF messages.  Currently the conditions for causing this problem
144are unknown.  Due to this, liss2ew should be treated as suspect.  Use at
145your own risk.  JMP 6-18-2003
146
147wave_serverV:
148appears not to reply to requests for a single sample of data. I noticed
149when testing wave_viewer, that if the start time and end time of a request were
150equal (in which case there should be one sample of data) then wave_serverV did
151not reply to the request (ASCII request) at all. No Data, No Flags, No Reply,
152No Nothing.  It needs to issue a reply to every request.
153DavidK 09/25/01
154
155Automatic restarts of adsend (using the "restartMe" line in the descriptor
156file) can cause an NT system to hang. Therefore, you should never
157use the autorestart feature with adsend, but you should bring down
158the entire Earthworm system if adsend needs to be restarted.
159LDD 5/31/2000 Comments added to adsend.desc, but leave this warning here!
160
161libsrc/utils/site.c: The strings used for station, channel and network are
162required to be fixed length with trailing spaces added to short names. If
163the strings given to site_index do not have these trailing blanks, SCN names
164will not match. This is not documented anywhere.  PNL 10/15/00
165
166
167KNOWN DEFICIENCIES:
168*******************
169ringtocoax:
170In Windows NT, the time resolution of sleep_ew() is about 16 msec (one clock
171tick).  On Solaris, the resolution is about 10 msec.  This is a problem for
172ringtocoax, since packet delays need to be set to a few milliseconds.
173
174threads functions: The KillThread function on WindowsNT and Solaris
175terminate the thread without ensuring that no mutexes are held. If a thread
176holds a mutex when it dies, no other thread can get that mutex. PNL 1/12/2000
177
178carlsubtrig:
179The system time must be set to GMT and ew_nt.cmd must have
180TZ=GMT for carlsubtrig to work.  Comments in ew_nt.cmd done 5/25/00. Barbara
181       
182localmag:
183needs GMT set on the system
184
185ew2seisvole:
186on NT, exits with horrible crash when system is stopped.
187
188NUMBER OF RINGS LIMITED ON SOLARIS:
189Under Solaris 2.6 (and probably other versions as well), the maximum number
190of shared memory segments is six. This means that on an out-of-the-box machine
191you can only configure six rings. If you try to configure more than that, you
192will see a cryptic message from tport_create about too many open files.  The
193fix to this problem is to add the following lines to the /etc/system
194file, and then reboot the system.
195
196 set shmsys:shminfo_shmmax = 4294967295
197 set shmsys:shminfo_shmmin = 1
198 set shmsys:shminfo_shmmni = 100
199 set shmsys:shminfo_shmseg = 20
200 set semsys:seminfo_semmns = 200
201 set semsys:seminfo_semmni = 70
202
203This allows for 20 rings.
204
205     Lucky Vidmar (7/6/2000)
206
207
208TRACEBUF messages.
209The definition of `endtime' of the TRACEBUF message is not documented.
210Some programmers are taking it as the "expected start time of the next
211TRACEBUF packet (if the sample interval is uniform.)" The more accepted
212practice is that `endtime' is the time of the last sample of the current
213TRACEBUF packet; that is, one sample interval less than the expected
214start time of the next TRACEBUF messsage. Using this last definition, if a
215TRACEBUF packet has exactly one sample, then its starttime and endtime are
216the identical. Clearly this distinction needs to be documented. The file
217waveform_format (in the /home/earthworm/DOC directory) gives no specifics
218about start or end times.  PNL 1/24/01
219
Note: See TracBrowser for help on using the repository browser.