Error: I get the following issue trying to start a database:
Starting database recovery or hot-backup restoration...
Database recovery in progress...
SEDNA Message: FATAL ERROR
in process [1207] /home/ubuntu/sedna/bin/se_sm xrdb
System error. This error means system malfunction.
Details: The data read from persistent storage is corrupt.
Cannot create data directory for soft fault logs
Cannot create soft fault log file
Partial from the event.log:
INFO 31/08/2011 18:07:14 (GOV pid=1147)
[gov_functions.cpp:log_out_system_information:87]: SEDNA version is 3.4.66
(64bit Release)
INFO 31/08/2011 18:07:14 (GOV pid=1147)
[gov_functions.cpp:log_out_system_information:95]: System: Linux
2.6.35-22-virtual x86_64
LOG 31/08/2011 18:07:14 (GOV pid=1147) [gov.cpp:main:239]: Process ping
server is ready
LOG 31/08/2011 18:07:14 (GOV pid=1147) [listener.cpp:client_listener:70]:
GOVERNOR is ready
SYS 31/08/2011 18:07:14 (GOV pid=1147) [usem.c:USemaphoreOpen:97]: semget
(code = 2): No such file or directory
LOG 31/08/2011 18:08:08 (SM xrdb pid=1207) [sm.cpp:main:493]: Event log is
ready
LOG 31/08/2011 18:08:08 (SM xrdb pid=1207) [sm.cpp:main:506]: Ping client
has been started
SYS 31/08/2011 18:08:08 (SM xrdb pid=1207)
[uprocess.c:uGetEnvironmentVariable:101]: getenv (code = 0): Success
SYS 31/08/2011 18:08:08 (SM xrdb pid=1207)
[uprocess.c:uGetEnvironmentVariable:101]: getenv (code = 0): Success
LOG 31/08/2011 18:08:08 (SM xrdb pid=1207)
[sm.cpp:recover_database_by_physical_and_logical_log:753]: Event log in
recovery procedure is ready
LOG 31/08/2011 18:08:08 (SM xrdb pid=1207)
[sm.cpp:recover_database_by_physical_and_logical_log:763]: Starting database
recovery or hot-backup restoration...
LOG 31/08/2011 18:08:08 (SM xrdb pid=1207)
[sm.cpp:recover_database_by_physical_and_logical_log:803]: Buffer manager is
started
LOG 31/08/2011 18:08:08 (SM xrdb pid=1207)
[sm.cpp:recover_database_by_physical_and_logical_log:810]: Database has been
recovered by physical log successfully
LOG 31/08/2011 18:08:08 (SM xrdb pid=1207)
[blk_mngmt.cpp:extend_tmp_file:629]: Temp file has been extended, size:
6400000
LOG 31/08/2011 18:08:08 (SM xrdb pid=1207)
[sm.cpp:recover_database_by_physical_and_logical_log:822]: Checkpoints are
disabled
LOG 31/08/2011 18:08:08 (SM xrdb pid=1207)
[sm.cpp:recover_database_by_physical_and_logical_log:845]: Wu is initialized
LOG 31/08/2011 18:08:08 (RCV xrdb pid=1211) [tr.cpp:TRmain:182]: Session
is ready
LOG 31/08/2011 18:08:13 (SM xrdb pid=1207)
[blk_mngmt.cpp:extend_data_file:594]: Data file has been extended, size:
b5410000
ERROR 31/08/2011 18:08:13 (SM xrdb pid=1207)
[wu.cpp:AnalyzePopularFailure:64]: PutBlockVersionToBuffer - invalid lxptr
(old version)
ERROR 31/08/2011 18:08:13 (SM xrdb pid=1207)
[wu.cpp:AnalyzePopularFailure:65]: PutBlockVersionToBuffer - xptr
0000000104ad0000, lxptr 0000000104ad0000
ERROR 31/08/2011 18:08:13 (SM xrdb pid=1207)
[wu.cpp:AnalyzePopularFailure:66]: PutBlockVersionToBuffer - vheader
{0000000101b40000, 0000000000010420}
Db create configuration:
se_cdb -bufs-num 3200 -max-trs-num 20 -data-file-init-size 2500 -upd-crt
.25 xrdb
free -t -m:
total used free shared buffers cached
Mem: 592 537 55 0 8 466
-/+ buffers/cache: 62 530
Swap: 0 0 0
Total: 592 537 55
Questions:
1. In the future, what kind of diagnostic information (memory, processes,
etc.) can I provide to help resolve issues? Are there specific linux
commands I should execute and provide results?
2. Is the root cause a Linux or Sedna configuration problem, or is the
problem a Sedna DB issue?
3. Is there any configuration that would be considered optimal for an
Amazon micro-instance that has nothing but Senda as the core process? The
DB can grow to 6 GB, number of required simultaneous connections is limited,
maybe 10 max. I don't mind recompiling code if that helps. (I think I have
asked this question before, sorry for being repetitive.)
4. I noticed that Sedna build is up to build 3.5.13. Are there any good
reasons to move from 3.4.66 to the latest build? Are there any stable
builds since the 3.4.66 release?
I have all the data associated with db stored elsewhere, so I am going to
simply rebuild the instance and move on. I just want to do my best from
having the issue occur again.
Any recommendations would be appreciated.
Thanks,
Malcolm
------------------------------------------------------------------------------
Special Offer -- Download ArcSight Logger for FREE!
Finally, a world-class log management solution at an even better
price-free! And you'll get a free "Love Thy Logs" t-shirt when you
download Logger. Secure your free ArcSight Logger TODAY!
http://p.sf.net/sfu/arcsisghtdev2dev
_______________________________________________
Sedna-discussion mailing list
Sedna-discussion@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sedna-discussion
