Informix - Problem description
| Problem IT41654 | Status: Closed |
MASSIVE PARALLEL ACTIVITY TO THE SAME TABLE CAN CREATE A BOTTEKNECK ON MUTEX WAIT PT_PARTNUM | |
| product: | |
INFORMIX SERVER / 5725A3900 / E10 - | |
| Problem description: | |
When we try to do a massive parallel activity on the same table
it can create a bottleneck "mutex wait pt_partnum"
IBM Informix Dynamic Server Version 14.10.FC8 -- On-Line -- Up 1
days 20:01:47 -- 239764 Kbytes
355 4eaf5d90 46d6e128 1 mutex wait
pt_1001e7 13cpu sqlexec
356 47c3fd90 46d74428 1 cond wait
netnorm 13cpu sqlexec
357 4faa1d48 46d64828 1 mutex wait
pt_1001e7 13cpu sqlexec
358 4ea94178 46d70e28 1 cond wait
netnorm 13cpu sqlexec
359 4ee6e0d0 46d6d828 1 running
9cpu sqlexec
360 4ee83808 46d63f28 1 mutex wait
pt_1001e7 13cpu sqlexec
361 4ee6c958 46d6bd28 1 mutex wait
pt_1001e7 13cpu sqlexec
362 4ee66a00 46d6ab28 1 mutex wait
pt_1001e7 13cpu sqlexec
363 4ee82178 46d73b28 1 mutex wait
pt_1001e7 13cpu sqlexec
364 4ee5f8b0 46d71728 1 cond wait
netnorm 12cpu sqlexec
365 4eebf028 46d60028 1 mutex wait
pt_1001e7 1cpu sqlexec
366 4ee93760 46d6a228 1 cond wait
netnorm 1cpu sqlexec
367 4ec5c028 46d69928 1 mutex wait
pt_1001e7 13cpu sqlexec
368 4ecfd610 46d63628 1 mutex wait
pt_1001e7 13cpu sqlexec
369 4ebd8d48 46d6cf28 1 cond wait
netnorm 13cpu sqlexec
370 4eb514c0 46d6b428 1 running
1cpu sqlexec
371 4eb99bf8 46d6f328 1 cond wait
netnorm 12cpu sqlexec
372 4eaec370 46d61b28 1 cond wait
netnorm 13cpu sqlexec
373 4eb72aa8 46d5f728 1 running
11cpu sqlexec
374 4ed1a220 46d73228 1 running
10cpu sqlexec
375 4ed52958 46d60928 1 mutex wait
pt_1001e7 13cpu sqlexec
376 506740d0 46d6c628 1 running
14cpu sqlexec
377 5024e808 46d72928 1 mutex wait
pt_1001e7 13cpu sqlexec
378 50331028 46d70528 1 cond wait
netnorm 1cpu sqlexec
379 5029a6b8 46d6ea28 1 cond wait
netnorm 12cpu sqlexec
380 502be028 46d65128 1 cond wait
netnorm 13cpu sqlexec
381 50620568 46d72028 1 running
12cpu sqlexec
382 5036cca0 46d6fc28 1 cond wait
netnorm 14cpu sqlexec
383 50361418 46d66328 1 mutex wait
pt_1001e7 13cpu sqlexec
384 51fdad00 46d61228 1 running
8cpu sqlexec
IBM Informix Dynamic Server Version 14.10.FC8 -- On-Line -- Up 1
days 20:01:49 -- 239764 Kbytes
355 4eaf5d90 46d6e128 1 mutex wait
pt_1001e6 10cpu sqlexec
356 47c3fd90 46d74428 1 mutex wait
pt_1001e6 8cpu sqlexec
357 4faa1d48 46d64828 1 mutex wait
pt_1001e6 1cpu sqlexec
358 4ea94178 46d70e28 1 mutex wait
pt_1001e6 8cpu sqlexec
359 4ee6e0d0 46d6d828 1 mutex wait
pt_1001e6 8cpu sqlexec
360 4ee83808 46d63f28 1 running
13cpu sqlexec
361 4ee6c958 46d6bd28 1 mutex wait
pt_1001e6 14cpu sqlexec
362 4ee66a00 46d6ab28 1 mutex wait
pt_1001e6 12cpu sqlexec
363 4ee82178 46d73b28 1 mutex wait
pt_1001e6 13cpu sqlexec
364 4ee5f8b0 46d71728 1 mutex wait
pt_1001e6 8cpu sqlexec
365 4eebf028 46d60028 1 mutex wait
pt_1001e6 8cpu sqlexec
366 4ee93760 46d6a228 1 running
11cpu sqlexec
367 4ec5c028 46d69928 1 mutex wait
pt_1001e6 1cpu sqlexec
368 4ecfd610 46d63628 1 mutex wait
pt_1001e6 13cpu sqlexec
369 4ebd8d48 46d6cf28 1 mutex wait
pt_1001e6 12cpu sqlexec
370 4eb514c0 46d6b428 1 mutex wait
pt_1001e6 10cpu sqlexec
371 4eb99bf8 46d6f328 1 mutex wait
pt_1001e6 1cpu sqlexec
372 4eaec370 46d61b28 1 running
9cpu sqlexec
373 4eb72aa8 46d5f728 1 mutex wait
pt_1001e6 8cpu sqlexec
374 4ed1a220 46d73228 1 mutex wait
pt_1001e6 12cpu sqlexec
375 4ed52958 46d60928 1 mutex wait
pt_1001e6 10cpu sqlexec
376 506740d0 46d6c628 1 mutex wait
pt_1001e6 8cpu sqlexec
377 5024e808 46d72928 1 mutex wait
pt_1001e6 8cpu sqlexec
378 50331028 46d70528 1 mutex wait
pt_1001e6 1cpu sqlexec
379 5029a6b8 46d6ea28 1 mutex wait
pt_1001e6 12cpu sqlexec
380 502be028 46d65128 1 mutex wait
pt_1001e6 1cpu sqlexec
381 50620568 46d72028 1 mutex wait
pt_1001e6 12cpu sqlexec
382 5036cca0 46d6fc28 1 mutex wait
pt_1001e6 10cpu sqlexec
383 50361418 46d66328 1 mutex wait
pt_1001e6 8cpu sqlexec
384 51fdad00 46d61228 1 mutex wait
pt_1001e6 10cpu sqlexec
I am able to reproduce on a table with (10) fragments and
without fragments.
I see larger no of mutex wait for a particular select query on a
single table with or without fragment.
for f in run*/*stk*;do awk '/mt_lock $/{i=1;next}i{print ;n++;
i--}END{print "Sum: " n}' $f|sort|uniq -c; echo; done
25 0x00000001001addc0 (oninit)isenter
23 0x00000001001af4cc (oninit)rspsclose
8 0x00000001001b53e0 (oninit)IPRA.$flfree
4 0x00000001001b6430 (oninit)get_leadpartp
10 0x00000001001ca44c (oninit)ptfree
9 0x00000001001cc3c4 (oninit)ptalloc
1 Sum: 79
11 0x00000001001addc0 (oninit)isenter
28 0x00000001001af4cc (oninit)rspsclose
14 0x00000001001b53e0 (oninit)IPRA.$flfree
11 0x00000001001b6430 (oninit)get_leadpartp
6 0x00000001001ca44c (oninit)ptfree
12 0x00000001001cc3c4 (oninit)ptalloc
1 Sum: 82
20 0x00000001001addc0 (oninit)isenter
27 0x00000001001af4cc (oninit)rspsclose
7 0x00000001001b53e0 (oninit)IPRA.$flfree
9 0x00000001001b6430 (oninit)get_leadpartp
4 0x00000001001ca44c (oninit)ptfree
16 0x00000001001cc3c4 (oninit)ptalloc
1 Sum: 83 | |
| Problem Summary: | |
**************************************************************** * USERS AFFECTED: * * Users of Informix Server prior to 14.10.FC10. * **************************************************************** * PROBLEM DESCRIPTION: * * See Error Description * **************************************************************** * RECOMMENDATION: * * Upgrade to Informix Server 14.10.FC10. * **************************************************************** | |
| Local Fix: | |
| Solution | |
| Workaround | |
**************************************************************** * USERS AFFECTED: * * Users of Informix Server prior to 14.10.FC10. * **************************************************************** * PROBLEM DESCRIPTION: * * See Error Description * **************************************************************** * RECOMMENDATION: * * Upgrade to Informix Server 14.10.FC10. * **************************************************************** | |
| Comment | |
Fixed in Informix Server 14.10.FC10. | |
| Timestamps | |
Date - problem reported : Date - problem closed : Date - last modified : | 01.08.2022 22.02.2023 22.02.2023 |
| Problem solved at the following versions (IBM BugInfos) | |
| Problem solved according to the fixlist(s) of the following version(s) | |