ER crash server
Roger set up ER from IDS 9.4.UC5 on Red Hat AS3 to IDS 7.31.UD2 and found that inserting many rows crashed the primary, with an mtex.c exception in thread CDRNsA22 and a panic; CDRSITES settings did not help. Neil Truby advised contacting IBM support. Superboer and Frank Qu suggested keeping transactions small. Roger said even ten or twenty rows in one statement crashed it. Unresolved.
Auto-generated by Claude from the posts below — may be imperfect; read the full thread.
Dear all:
I do a ER primary-target between 2 server:
primary server : redhat as3 , ids 9.4.uc5
secondary server : redhat 7.2 , ids7.31.ud2
sqlhosts's set up:
cdr1 group - -
i=11
ids94 onsoctcp 10.35.7.3 1536 g=cdr1
cdr2 group - -
i=22
ids73 onsoctcp 10.35.7.1 1526 g=cdr2
the process that setup list below:
cdr define server -c cdr1 -I cdr1
cdr define server -c cdr2 -I cdr2 -S cdr1
cdr define repl -c cdr1 -C ignore -A repl1 \\
"P test@cdr1:informix.t1" "select * from t1" \\
"R test@cdr2:informix.t1" "select * from t1"
cdr start repl repl1
everything seem going fine. I insert one or two rows on primary serve,
it did replicated to the secondary server. But if I insert a lots of
rows of data continuosly , It will crash the primary server.
the error message from online.log:
10:41:08 Who: Session(51, informix@redhat72, 1361, 0x44d8b190)
Thread(61, CDRNsA22, 44d5db80, 1)
File: mtex.c Line: 431
10:41:08 Results: Exception Caught. Type: MT_EX_OS, Context: mem
10:41:08 Action: Please notify IBM Informix Technical Support.
10:41:08 stack trace for pid 2111 written to /tmp/af.4257f04
10:41:08 See Also: /tmp/af.4257f04, shmem.4257f04.0
10:41:13 mtex.c, line 431, thread 61, proc id 2111, No Exception
Handler.
10:41:13 The Master Daemon Died
10:41:14 PANIC: Attempting to bring system downI Even set CDRSITES_92X="22" or CDRSITS_731="22", It won't work.
Anyone can help me out of this?
<roger@star2000.com.tw> wrote in message
news:1153884225.575169.237680@m73g2000cwd.googlegroups.com...
> 10:41:08 Who: Session(51, informix@redhat72, 1361, 0x44d8b190)
> Thread(61, CDRNsA22, 44d5db80, 1)
> File: mtex.c Line: 431
> 10:41:08 Results: Exception Caught. Type: MT_EX_OS, Context: mem
> 10:41:08 Action: Please notify IBM Informix Technical Support.
> 10:41:08 stack trace for pid 2111 written to /tmp/af.4257f04
> 10:41:08 See Also: /tmp/af.4257f04, shmem.4257f04.0
> 10:41:13 mtex.c, line 431, thread 61, proc id 2111, No Exception
> Handler.
> 10:41:13 The Master Daemon Died
> 10:41:14 PANIC: Attempting to bring system down> I Even set CDRSITES_92X="22" or CDRSITS_731="22", It won't work.
> Anyone can help me out of this?
It looks like the sort of thing you need to contact IBM Informix Tech
Support about, Roger.
> But if I insert a lots of
> rows of data continuosly , It will crash the primary server.
> the error message from online.log:
For V7 ; pre 9.4 engines and ER keep trx small since a lot has to be
done in memory
so after x (say 100 recs ) do a commit.
very thin ice: may be i am way way off here;
in V9.4 and later there is some sort of overflow to process big trx...
Madison please comment.
Superboer.
BTW it may help to post also the stacktrace and contact TS.
roger@star2000.com.tw schreef:
> Dear all:
> I do a ER primary-target between 2 server:
> primary server : redhat as3 , ids 9.4.uc5
> secondary server : redhat 7.2 , ids7.31.ud2
> sqlhosts's set up:
> cdr1 group - -
> i=11
> ids94 onsoctcp 10.35.7.3 1536 g=cdr1
> cdr2 group - -
> i=22
> ids73 onsoctcp 10.35.7.1 1526 g=cdr2
>
> the process that setup list below:
> cdr define server -c cdr1 -I cdr1
> cdr define server -c cdr2 -I cdr2 -S cdr1
> cdr define repl -c cdr1 -C ignore -A repl1 \\
> "P test@cdr1:informix.t1" "select * from t1" \\
> "R test@cdr2:informix.t1" "select * from t1"
> cdr start repl repl1
>
> everything seem going fine. I insert one or two rows on primary serve,
> it did replicated to the secondary server. But if I insert a lots of
> rows of data continuosly , It will crash the primary server.
> the error message from online.log:
> 10:41:08 Who: Session(51, informix@redhat72, 1361, 0x44d8b190)
> Thread(61, CDRNsA22, 44d5db80, 1)
> File: mtex.c Line: 431
> 10:41:08 Results: Exception Caught. Type: MT_EX_OS, Context: mem
> 10:41:08 Action: Please notify IBM Informix Technical Support.
> 10:41:08 stack trace for pid 2111 written to /tmp/af.4257f04
> 10:41:08 See Also: /tmp/af.4257f04, shmem.4257f04.0
> 10:41:13 mtex.c, line 431, thread 61, proc id 2111, No Exception
> Handler.
> 10:41:13 The Master Daemon Died
> 10:41:14 PANIC: Attempting to bring system down> I Even set CDRSITES_92X="22" or CDRSITS_731="22", It won't work.
> Anyone can help me out of this?
My rule of thumb is,
Never replicate large volume of data!
You should split it into smaller transactions, e.g, each only insert a
couple hundred rows, or just manually insert the large volume of data at
each site ( " begin work without replication" I did this lot may
manually synchronization).
Frank
Superboer wrote:
>> But if I insert a lots of
>> rows of data continuosly , It will crash the primary server.
>>the error message from online.log:
>>
>>
>
>For V7 ; pre 9.4 engines and ER keep trx small since a lot has to be
>done in memory
>
>so after x (say 100 recs ) do a commit.
>
>very thin ice: may be i am way way off here;
>
>in V9.4 and later there is some sort of overflow to process big trx...
>
>
>Madison please comment.
>
>Superboer.
>
>BTW it may help to post also the stacktrace and contact TS.
>
>
>roger@star2000.com.tw schreef:
>
>
>
>>Dear all:
>> I do a ER primary-target between 2 server:
>> primary server : redhat as3 , ids 9.4.uc5
>> secondary server : redhat 7.2 , ids7.31.ud2
>>sqlhosts's set up:
>>cdr1 group - -
>>i=11
>>ids94 onsoctcp 10.35.7.3 1536 g=cdr1
>>cdr2 group - -
>> i=22
>>ids73 onsoctcp 10.35.7.1 1526 g=cdr2
>>
>>the process that setup list below:
>> cdr define server -c cdr1 -I cdr1
>>cdr define server -c cdr2 -I cdr2 -S cdr1
>>cdr define repl -c cdr1 -C ignore -A repl1 \\
>> "P test@cdr1:informix.t1" "select * from t1" \\
>> "R test@cdr2:informix.t1" "select * from t1"
>>cdr start repl repl1
>>
>>everything seem going fine. I insert one or two rows on primary serve,
>>it did replicated to the secondary server. But if I insert a lots of
>> rows of data continuosly , It will crash the primary server.
>>the error message from online.log:
>> 10:41:08 Who: Session(51, informix@redhat72, 1361, 0x44d8b190)
>> Thread(61, CDRNsA22, 44d5db80, 1)
>> File: mtex.c Line: 431
>>10:41:08 Results: Exception Caught. Type: MT_EX_OS, Context: mem
>>10:41:08 Action: Please notify IBM Informix Technical Support.
>>10:41:08 stack trace for pid 2111 written to /tmp/af.4257f04
>>10:41:08 See Also: /tmp/af.4257f04, shmem.4257f04.0
>>10:41:13 mtex.c, line 431, thread 61, proc id 2111, No Exception
>>Handler.
>>10:41:13 The Master Daemon Died
>>10:41:14 PANIC: Attempting to bring system down>>I Even set CDRSITES_92X="22" or CDRSITS_731="22", It won't work.
>>Anyone can help me out of this?
>>
>>
>
>_______________________________________________
>Informix-list mailing list
>Informix-list@iiug.org
>http://www.iiug.org/mailman/listinfo/informix-list
>
>
>
--
Yunyao "Frank" Qu
Computer Sciences Corporation(CSC)
NOAA/CLASS, (301)817-4696
Yunyao (Frank) Qu 寫道：
> My rule of thumb is,
>
> Never replicate large volume of data!
>
> You should split it into smaller transactions, e.g, each only insert a
> couple hundred rows, or just manually insert the large volume of data at
> each site ( " begin work without replication" I did this lot may
> manually synchronization).
>
> Frank
>
>
> Superboer wrote:
>
> >> But if I insert a lots of
> >> rows of data continuosly , It will crash the primary server.
> >>the error message from online.log:
> >>
> >>
> >
> >For V7 ; pre 9.4 engines and ER keep trx small since a lot has to be
> >done in memory
> >
> >so after x (say 100 recs ) do a commit.
> >
> >very thin ice: may be i am way way off here;
> >
> >in V9.4 and later there is some sort of overflow to process big trx...
> >
> >
> >Madison please comment.
> >
> >Superboer.
> >
> >BTW it may help to post also the stacktrace and contact TS.
> >
> >
> >roger@star2000.com.tw schreef:
> >
> >
> >
> >>Dear all:
> >> I do a ER primary-target between 2 server:
> >> primary server : redhat as3 , ids 9.4.uc5
> >> secondary server : redhat 7.2 , ids7.31.ud2
> >>sqlhosts's set up:
> >>cdr1 group - -
> >>i=11
> >>ids94 onsoctcp 10.35.7.3 1536 g=cdr1
> >>cdr2 group - -
> >> i=22
> >>ids73 onsoctcp 10.35.7.1 1526 g=cdr2
> >>
> >>the process that setup list below:
> >> cdr define server -c cdr1 -I cdr1
> >>cdr define server -c cdr2 -I cdr2 -S cdr1
> >>cdr define repl -c cdr1 -C ignore -A repl1 \\
> >> "P test@cdr1:informix.t1" "select * from t1" \\
> >> "R test@cdr2:informix.t1" "select * from t1"
> >>cdr start repl repl1
> >>
> >>everything seem going fine. I insert one or two rows on primary serve,
> >>it did replicated to the secondary server. But if I insert a lots of
> >> rows of data continuosly , It will crash the primary server.
> >>the error message from online.log:
> >> 10:41:08 Who: Session(51, informix@redhat72, 1361, 0x44d8b190)
> >> Thread(61, CDRNsA22, 44d5db80, 1)
> >> File: mtex.c Line: 431
> >>10:41:08 Results: Exception Caught. Type: MT_EX_OS, Context: mem
> >>10:41:08 Action: Please notify IBM Informix Technical Support.
> >>10:41:08 stack trace for pid 2111 written to /tmp/af.4257f04
> >>10:41:08 See Also: /tmp/af.4257f04, shmem.4257f04.0
> >>10:41:13 mtex.c, line 431, thread 61, proc id 2111, No Exception
> >>Handler.
> >>10:41:13 The Master Daemon Died
> >>10:41:14 PANIC: Attempting to bring system down> >>I Even set CDRSITES_92X="22" or CDRSITS_731="22", It won't work.
> >>Anyone can help me out of this?
> >>
> >>
> >
> >_______________________________________________
> >Informix-list mailing list
> >Informix-list@iiug.org
> >http://www.iiug.org/mailman/listinfo/informix-list
> >
> >
> >
>
> --
> Yunyao "Frank" Qu
> Computer Sciences Corporation(CSC)
> NOAA/CLASS, (301)817-4696
Thanks for all of your's help.
I said that a lot of rows replicated , It's just ten or twenty rows
that insert in a SQL statement will crash the primary server .
roger