Segmentation Fault after upgrade 2.1.3 -> 2.1.4 [CORE3409] #3773
Comments
|
Commented by: @pmakowski why don't you use the Fedora package ? |
|
Commented by: @dyemanov All you need to do is to provide us (the developers) with a crash dump. See here: |
|
Commented by: Fernando Guidorizzi (mephisty) Philippe, i was using the 2.1.3 rpm from http://firebirdsql.org Disabling sweep doesnt work, I have a 'agressive' DefaultDbCachePages with 50000, changed to 5000 ... |
|
Commented by: @pmakowski before switching to fedora package, provide the crash dum, it may be useful IMHO DefaultDbCachePages with 50000 under Linux is really not necessary |
|
Commented by: Fernando Guidorizzi (mephisty) Dmitry, im waiting the next segfault and will attach the dump |
|
Commented by: Fernando Guidorizzi (mephisty) Before BugCheckAbort =1 and DefaultDbCachePages=5000 Waiting to get the dump file... |
Modified by: @AlexPeshkoffassignee: Alexander Peshkov [ alexpeshkoff ] |
|
Commented by: Fernando Guidorizzi (mephisty) Could i post a compressed 15mb crash dump here? Edited: |
|
Commented by: Fernando Guidorizzi (mephisty) Sorry, im unable to install debuginfo. Just the core dump could help? |
|
Commented by: @AlexPeshkoff Are you using SF binaries? If yes - can. Upload it here as an attachment. |
|
Commented by: Fernando Guidorizzi (mephisty) Crash Dump With SF binaries. Sorry 7z format. |
Modified by: Fernando Guidorizzi (mephisty)Attachment: core.13455.7z [ 11921 ] |
|
Commented by: @AlexPeshkoff 7z not a problem |
|
Commented by: @pmakowski Alex you mean 2.1.4.18393-0 from there http://firebirdsql.org/index.php?op=files&id=engine_214 |
|
Commented by: Fernando Guidorizzi (mephisty) Alex, tks! |
|
Commented by: Fernando Guidorizzi (mephisty) Im waiting to access the production server. and get thisgdb /opt/firebird/bin/.debug/fbserver.debug core.13455 warning: core file may not match specified executable file. warning: .dynamic section for "/lib/libncurses.so.5" is not at the expected address (wrong library or version mismatch?) warning: .dynamic section for "/lib/libpthread.so.0" is not at the expected address (wrong library or version mismatch?) warning: .dynamic section for "/lib/libgcc_s.so.1" is not at the expected address (wrong library or version mismatch?) warning: .dynamic section for "/lib/ld-linux.so.2" is not at the expected address (wrong library or version mismatch?) warning: .dynamic section for "/lib/libtinfo.so.5" is not at the expected address (wrong library or version mismatch?) If could help.... |
|
Commented by: Fernando Guidorizzi (mephisty) Production server Core dump .... [root@localhost .debug]# gdb /opt/firebird/bin/.debug/fbserver.debug core.1725 warning: core file may not match specified executable file. Thread 14 (Thread 1725): Thread 13 (Thread 1741): Thread 12 (Thread 1728): Thread 11 (Thread 2194): Thread 10 (Thread 1739): Thread 9 (Thread 2193): Thread 8 (Thread 1811): Thread 7 (Thread 1809): Thread 6 (Thread 1810): Thread 5 (Thread 1740): Thread 4 (Thread 1797): Thread 3 (Thread 1794): Thread 2 (Thread 1738): Thread 1 (Thread 1732): |
|
Commented by: @AlexPeshkoff This is what was needed. Please keep that core dump together with debuginfo - may be some more info will be needed. |
|
Commented by: Fernando Guidorizzi (mephisty) Maybe related with ticket |
|
Commented by: Fernando Guidorizzi (mephisty) Alex! tks!!! |
|
Commented by: @AlexPeshkoff Fernando, please rename /opt/firebird with 2.5 to something else and keep core in it - if it will become required, we will get required info from it. |
|
Commented by: @AlexPeshkoff Fernando - one more question. Is it hard to reproduce a bug? I.e. can you provide us with a database and a scipt (or some program) running which say during a day the bug will be reproduced. |
|
Commented by: Fernando Guidorizzi (mephisty) Alexander, just to confirm 2.5? will coexist with 2.1.3? The bug occurred about 20 times yesterday, in random querys and time intervals. Tomorrow I will restore the 2.1.4 version and try to reproduce. Tks! |
|
Commented by: @AlexPeshkoff About coexistance of 2.5 and 2.1.3 - it's offtopic here in the tracker, please contact me privately or in fb-devel. That's nice (for us) that bug is easy to reproduce. But can you provide a database to try to reproduce at developers box? (Do not care much about size - main problem is does it contatin critical data.) |
|
Commented by: Fernando Guidorizzi (mephisty) I cant provide the database, sorry. |
|
Commented by: @AlexPeshkoff Understandable thing with database. What about SSD - if 2.1.3 works fine with your SSD - that's not a problem. |
|
Commented by: Fernando Guidorizzi (mephisty) I tried to reproduce with no success. is there a scenario where a non-updated client library could cause a SegFault? |
|
Commented by: Fernando Guidorizzi (mephisty) Vlad, my unique confirmation was in the FlameRobin reporting 2.1.4 I will uninstall the 2.1.3 and rename firebird folder to make sure. |
|
Commented by: Fernando Guidorizzi (mephisty) 2.1.4 package already did the job ..archiving firebird folder.... |
|
Commented by: Fernando Guidorizzi (mephisty) New core dump. root@localhost .debug]# gdb fbserver.debug core.30557 warning: core file may not match specified executable file. Thread 16 (Thread 21014): Thread 15 (Thread 20881): Thread 14 (Thread 30557): Thread 13 (Thread 21015): Thread 12 (Thread 20880): #17 0x08142672 in EXE_receive (tdbb=0x49d9c70, request=0xb3f983fc, msg=1, length=263, buffer=0xacbe4aa4 "", top_level=true) Thread 11 (Thread 30567): Thread 10 (Thread 5087): Thread 9 (Thread 5088): Thread 8 (Thread 5085): Thread 7 (Thread 20879): Thread 6 (Thread 30560): Thread 5 (Thread 5086): Thread 4 (Thread 21016): Thread 3 (Thread 4788): Thread 2 (Thread 21013): Thread 1 (Thread 20882): #1 0x08109f2f in LocksCacheJrd::CachedLock\::get (this=0xb6f8edd4, tdbb=0x67dcc70, key=0x67da7a0 "Q\200\a") |
|
Commented by: Fernando Guidorizzi (mephisty) Philippe, Probably, i have isolated the problematic table. |
|
Commented by: Fernando Guidorizzi (mephisty) I cant reproduce the workload to generate the error in a test server. Any ideas? Tks |
|
Commented by: @hvlad Classic shouldn't have this issue. |
|
Commented by: Fernando Guidorizzi (mephisty) i was making a backup (gbak), and the following error occurred fbserver[4623]: segfault at 4c1c3b0d ip 009f26d7 sp b65e8424 error 4 in http://libpthread-2.13.so[9e9000+17000] And in a random way this on gbak "ERROR:gds_$open_blob failed" But happened in both versions(2.1.3, 2.1.4). Could my database are corrupted and generating this main issue? |
|
Commented by: @AlexPeshkoff Your database appears corrupted (did you try gfix?), but this should not be a reason for segfaults. |
|
Commented by: Fernando Guidorizzi (mephisty) I will gfix&gbak the production database...and try the 2.1.4 |
|
Commented by: Fernando Guidorizzi (mephisty) I did the gfix, and repaired the problems encoutered. I tried ;/ Thanks! to everyone that was involved |
|
Commented by: @hvlad Could you try Classic Server ? |
|
Commented by: Fernando Guidorizzi (mephisty) Yes, tomorrow i return with the result of the test |
|
Commented by: Fernando Guidorizzi (mephisty) This weekend, i will do the classic test |
|
Commented by: Fernando Guidorizzi (mephisty) Im unable to install classic version, The server was rejecting the incoming connections and returning with the message I searched for some solution, not found anything.... Any help?! tks! |
|
Commented by: @AlexPeshkoff Tracker is not good place for discussing install problems. Please contact me privately, peshkoff at mail dot ru, to solve your problems. |
Modified by: @hvladassignee: Alexander Peshkov [ alexpeshkoff ] => Vlad Khorsun [ hvlad ] |
Modified by: @hvladstatus: Open [ 1 ] => Resolved [ 5 ] resolution: Fixed [ 1 ] Fix Version: 2.1.5 [ 10420 ] |
Modified by: @pcisarstatus: Resolved [ 5 ] => Closed [ 6 ] |
Submitted by: Fernando Guidorizzi (mephisty)
Is duplicated by CORE3655
Attachments:
core.13455.7z
After a upgrade to newest version.
The following msg was received...........
fbserver[12394]: segfault at aa121000 ip 0027e6d7 sp 07aaa668 error 4 in http://libc-2.12.2.so[150000+184000]
fbserver[13238]: segfault at aa918000 ip 0047d6f5 sp 06735688 error 4 in http://libc-2.12.2.so[34f000+184000]
localhost.localdomain (Client) Thu Mar 24 08:21:35 2011
/opt/firebird/bin/fbguard: /opt/firebird/bin/fbserver terminated abnormally (-1)
localhost.localdomain (Client) Thu Mar 24 08:21:35 2011
/opt/firebird/bin/fbguard: guardian starting bin/fbserver
Now, i disabled the Sweep Interval ....
Any ideas???
Tks in advance!
Commits: a406f32
The text was updated successfully, but these errors were encountered: