Re: Logical Logs in Enterprise Replication
- From: Madison Pruet <mpruet1@xxxxxxxxxxx>
- Date: Wed, 30 May 2007 00:41:18 GMT
Navaid Arif wrote:
Hi All,
I am using Enterprise Replication on IDS10 UC5. I have 33 Logical Logs for IDS, each of 5000 pages i.e. 20,000KB.
Test >onstat -l
IBM Informix Dynamic Server Version 10.00.UC5 -- On-Line --
Up 09:00:11 -- 81408 Kbytes
Physical Logging
Buffer bufused bufsize numpages numwrits pages/io
P-2 5 128 1303 104 12.53
phybegin physize phypos phyused %used
1:263 2500 497 5 0.20
Logical Logging
Buffer bufused bufsize numrecs numpages numwrits recs/pages pages/io
L-1 0 128 90892 18432 2667 4.9 6.9
Subsystem numrecs Log Space used
OLDRSAM 81180 43897416
SBLOB 9706 22230368
CDR 6 296
address number flags uniqid begin size used %used
502ac4b0 1 U-B---- 67 1:2763 5000 5000 100.00
502ac4f8 2 U-B---- 68 1:7763 5000 5000 100.00
502ac540 3 U-B---- 69 1:12763 5000 5000 100.00
502ac588 4 U-B---- 70 1:17763 5000 5000 100.00
502ac5d0 5 U-B---- 71 1:22763 5000 5000 100.00
502ac618 6 U-B---- 72 1:27763 5000 5000 100.00
502ac660 7 U-B---- 73 1:32763 5000 5000 100.00
502ac6a8 8 U---C-L 74 1:37763 5000 355 7.10
502ac6f0 9 U-B---- 42 1:42763 5000 5000 100.00
502ac738 10 U-B---- 43 1:47763 5000 5000 100.00
502ac780 11 U-B---- 44 1:52763 5000 5000 100.00
502ac7c8 12 U-B---- 45 1:57763 5000 5000 100.00
502ac810 13 U-B---- 46 1:62763 5000 5000 100.00
502ac858 14 U-B---- 47 1:67763 5000 5000 100.00
502ac8a0 15 U-B---- 48 1:72763 5000 5000 100.00
502ac8e8 16 U-B---- 49 1:77763 5000 5000 100.00
502ac930 17 U-B---- 50 1:82763 5000 5000 100.00
502ac978 18 U-B---- 51 1:87763 5000 5000 100.00
502ac9c0 19 U-B---- 52 1:92763 5000 5000 100.00
502aca08 20 U-B---- 53 1:97763 5000 5000 100.00
502aca50 21 U-B---- 54 1:102763 5000 5000 100.00
502aca98 22 U-B---- 55 1:107763 5000 5000 100.00
502acae0 23 U-B---- 56 1:112763 5000 5000 100.00
502acb28 24 U-B---- 57 1:117763 5000 5000 100.00
502acb70 25 U-B---- 58 1:122763 5000 5000 100.00
502acbb8 26 U-B---- 59 1:127763 5000 5000 100.00
502acc00 27 U-B---- 60 1:132763 5000 5000 100.00
502acc48 28 U-B---- 61 1:137763 5000 5000 100.00
502acc90 29 U-B---- 62 1:142763 5000 5000 100.00
502accd8 30 U-B---- 63 1:147763 5000 5000 100.00
502acd20 31 U-B---- 64 1:155688 5000 5000 100.00
502acd68 32 U-B---- 65 1:160688 5000 5000 100.00
502acdb0 33 U-B---- 66 1:165688 5000 5000 100.00
33 active, 33 total
Test >
During update of almost 20,000 records, I am facing following messages in ONLINE.LOG.
18:54:33 Maximum server connections 13
18:59:33 Fuzzy Checkpoint Completed: duration was 0 seconds, 2
buffers not flushed.
18:59:33 Checkpoint loguniq 71, logpos 0x83e060, timestamp:
0x16d4e7f
18:59:33 Maximum server connections 13
19:04:34 Fuzzy Checkpoint Completed: duration was 0 seconds, 8
buffers not flushed.
19:04:34 Checkpoint loguniq 71, logpos 0x854960, timestamp:
0x16d52db
19:04:34 Maximum server connections 13
19:09:34 Fuzzy Checkpoint Completed: duration was 0 seconds,
16 buffers not flushed.
19:09:34 Checkpoint loguniq 71, logpos 0x86f140, timestamp:
0x16d56af
19:09:34 Maximum server connections 13
19:11:04 dynamically allocated 2000 locks
19:11:07 WARNING: Buffer pool size is too small for read ahead
operation
19:11:15 dynamically allocated 4000 locks
19:11:21 Logical Log 71 Complete, timestamp: 0x16dbb35.
19:11:47 Warning - The log space appears to be configured too small
for use with Enterprise Replication (ER). ER may require
additional logical logs to avoid a DDRBLOCK state
and/or replay position log wrap. It is recommended
that the
logical log configuration be expanded.
19:12:14 dynamically allocated 8000 locks
19:12:19 Logical Log 72 Complete, timestamp: 0x16e73e7.
19:12:34 CDR NIF receive failed asfcode: -25582 oserr 0
19:12:35 CDR connection to server lost, id 1, name <grpdr>
Reason: connection lost
19:13:52 Dynamically allocated new virtual shared memory
segment (size 8192KB)
19:14:46 Fuzzy Checkpoint Completed: duration was 3 seconds,
1132 buffers not flushed.
19:14:46 Checkpoint loguniq 73, logpos 0x1371770, timestamp:
0x16ee46b
19:14:46 Maximum server connections 13
19:17:58 CDR: Re-connected to server, id 1, name <grpdr>
19:19:46 Fuzzy Checkpoint Completed: duration was 0 seconds,
1135 buffers not flushed.
19:19:46 Checkpoint loguniq 73, logpos 0x13817a0, timestamp:
0x16ee7b2
19:19:46 Maximum server connections 13
19:20:48 Logical Log 73 Complete, timestamp: 0x16ee9b7.
19:24:46 Fuzzy Checkpoint Completed: duration was 0 seconds,
1140 buffers not flushed.
19:24:46 Checkpoint loguniq 74, logpos 0xfb7f0, timestamp:
0x16ef16d
19:24:46 Maximum server connections 13
19:29:49 Fuzzy Checkpoint Completed: duration was 3 seconds, 9
buffers not flushed.
19:29:49 Checkpoint loguniq 74, logpos 0x1040d0, timestamp:
0x16ef478
19:29:49 Maximum server connections 13
19:33:29 Dynamically allocated new virtual shared memory
segment (size 8192KB)
19:34:49 Fuzzy Checkpoint Completed: duration was 0 seconds, 6
buffers not flushed.
19:34:49 Checkpoint loguniq 74, logpos 0x10f3a0, timestamp:
0x16ef781
Can anyone plz tell me what is the reason for this warning message if I have large pool for Logical Logs.
We internally are calculating the rate of advancing the replay position with the rate of advancing the current log position. Based on the current rates of change of the current and replay position it appears that we could run into a log wrap within an hour, then we print the warning. You might see this in the case where you where you ran a heavy load, or maybe a larger than usual updating transaction.
Regards,.
*Navaid Arif*
- References:
- Logical Logs in Enterprise Replication
- From: Navaid Arif
- Logical Logs in Enterprise Replication
- Prev by Date: RE: Migration advise request SE or IDS Express
- Next by Date: Re: Informix 7.30 (I know!)
- Previous by thread: Logical Logs in Enterprise Replication
- Next by thread: Re: THREE EXCELLENT QA ANALYST'S AVAILABLE
- Index(es):
Relevant Pages
|
|
