ER performance

IDS11.50 FC8, Redhat Linux. Three sites ER update anywhere. Apparently, we have one site ER constantly queuing data locally, which can not replicate data to another site fast enough. The output of the cdr list server and onstat -g nif are as following ( in the site where ER queues data). The "BLOCK" in th line, 22 g_ncdcops RUN,BLOCK 9 220847207 111623315 means, sLow network? slow server?, or .... ? (The other two sites do not have "BLOCK" there.) More comments? Thanks, Frank [informix@maggie ~]$ cdr list serve SERVER ID STATE STATUS QUEUE CONNECTION CHANGED ----------------------------------------------------------------------- g_ncdcops 22 Active Connected 710147964 Nov 15 20:06:57 g_ngdcops 23 Active Connected 219989883 Nov 15 21:24:32 g_nsofops 21 Active Local 0 [informix@maggie ~]$ onstat -g nif IBM Informix Dynamic Server Version 11.50.FC8 -- On-Line -- Up 26 days 05:41:21 -- 4966528 KbytesNIF anchor Block: 1131a77d0 nifGState RUN RetryTimeout 300 CDR connections: Id Name State Version Sent Received --------------------------------------------------------------------------- 22 g_ncdcops RUN,BLOCK 9 220847207 111623315 23 g_ngdcops RUN 9 231228996 122563913 --0015175cfed4bf019e04b1cd2e4d

The block simply means that it is processing a large transaction. The receive queue is fairly full and a block is sent to the source nodes to let them know not to send anything until the receive queue has been processed. This could be the result of a large object such as a blob being transmitted to the target node. From: "FRANK" <yunyaoqu@gmail.com> To: ids@iiug.org Date: 11/15/2011 04:06 PM Subject: ER performance [25429] Sent by: ids-bounces@iiug.org IDS11.50 FC8, Redhat Linux. Three sites ER update anywhere. Apparently, we have one site ER constantly queuing data locally, which can not replicate data to another site fast enough. The output of the cdr list server and onstat -g nif are as following ( in the site where ER queues data). The "BLOCK" in th line, 22 g_ncdcops RUN,BLOCK 9 220847207 111623315 means, sLow network? slow server?, or .... ? (The other two sites do not have "BLOCK" there.) More comments? Thanks, Frank [informix@maggie ~]$ cdr list serve SERVER ID STATE STATUS QUEUE CONNECTION CHANGED ----------------------------------------------------------------------- g_ncdcops 22 Active Connected 710147964 Nov 15 20:06:57 g_ngdcops 23 Active Connected 219989883 Nov 15 21:24:32 g_nsofops 21 Active Local 0 [informix@maggie ~]$ onstat -g nif IBM Informix Dynamic Server Version 11.50.FC8 -- On-Line -- Up 26 days 05:41:21 -- 4966528 KbytesNIF anchor Block: 1131a77d0 nifGState RUN RetryTimeout 300 CDR connections: Id Name State Version Sent Received --------------------------------------------------------------------------- 22 g_ncdcops RUN,BLOCK 9 220847207 111623315 23 g_ngdcops RUN 9 231228996 122563913 --0015175cfed4bf019e04b1cd2e4d ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

Thanks, Madison! I rerun the command in the local box. see below. The line, 22 g_ncdcops RUN,BLOCK 9 221792052 111939781 means, who is processing the big jobs? the local ER or remote g_ncdcops ? Can slow/busy network cause this issue? Thanks, Frank [informix@maggie info]$ onstat -g nif IBM Informix Dynamic Server Version 11.50.FC8 -- On-Line -- Up 26 days 07:27:02 -- 4966528 KbytesNIF anchor Block: 1131a77d0 nifGState RUN RetryTimeout 300 CDR connections: Id Name State Version Sent Received --------------------------------------------------------------------------- 22 g_ncdcops RUN,BLOCK 9 221792052 111939781 23 g_ngdcops RUN 9 231884869 122563914 On Tue, Nov 15, 2011 at 5:11 PM, Madison Pruet <mpruet@us.ibm.com> wrote: > The block simply means that it is processing a large transaction. The > receive queue is fairly full and a block is sent to the source nodes to let > them know not to send anything until the receive queue has been processed. > This could be the result of a large object such as a blob being transmitted > to the target node. > > From: "FRANK" <yunyaoqu@gmail.com> > > To: ids@iiug.org > > Date: 11/15/2011 04:06 PM > > Subject: ER performance [25429] > > Sent by: ids-bounces@iiug.org > > IDS11.50 FC8, Redhat Linux. Three sites ER update anywhere. > > Apparently, we have one site ER constantly queuing data locally, which can > not replicate data to another site fast enough. > > The output of the cdr list server and onstat -g nif are as following > ( in the site where ER queues data). > > The "BLOCK" in th line, > > 22 g_ncdcops RUN,BLOCK 9 220847207 111623315 > > means, sLow network? slow server?, or .... ? (The other two sites do not > have "BLOCK" there.) > > More comments? > > Thanks, > Frank > > [informix@maggie ~]$ cdr list serve > SERVER ID STATE STATUS QUEUE CONNECTION CHANGED > ----------------------------------------------------------------------- > g_ncdcops 22 Active Connected 710147964 Nov 15 20:06:57 > g_ngdcops 23 Active Connected 219989883 Nov 15 21:24:32 > g_nsofops 21 Active Local 0 > > [informix@maggie ~]$ onstat -g nif > IBM Informix Dynamic Server Version 11.50.FC8 -- On-Line -- Up 26 days > 05:41:21 -- 4966528 Kbytes> NIF anchor Block: 1131a77d0 > > nifGState RUN > > RetryTimeout 300 > CDR connections: > Id Name State Version Sent Received > --------------------------------------------------------------------------- > > 22 g_ncdcops RUN,BLOCK 9 220847207 111623315 > 23 g_ngdcops RUN 9 231228996 122563913 > > --0015175cfed4bf019e04b1cd2e4d > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --000e0cdfd67654a4ad04b1ce7569

The network has nothing to do with it. The issue is with the apply. From: "FRANK" <yunyaoqu@gmail.com> To: ids@iiug.org Date: 11/15/2011 05:38 PM Subject: Re: ER performance [25431] Sent by: ids-bounces@iiug.org Thanks, Madison! I rerun the command in the local box. see below. The line, 22 g_ncdcops RUN,BLOCK 9 221792052 111939781 means, who is processing the big jobs? the local ER or remote g_ncdcops ? Can slow/busy network cause this issue? Thanks, Frank [informix@maggie info]$ onstat -g nif IBM Informix Dynamic Server Version 11.50.FC8 -- On-Line -- Up 26 days 07:27:02 -- 4966528 KbytesNIF anchor Block: 1131a77d0 nifGState RUN RetryTimeout 300 CDR connections: Id Name State Version Sent Received --------------------------------------------------------------------------- 22 g_ncdcops RUN,BLOCK 9 221792052 111939781 23 g_ngdcops RUN 9 231884869 122563914 On Tue, Nov 15, 2011 at 5:11 PM, Madison Pruet <mpruet@us.ibm.com> wrote: > The block simply means that it is processing a large transaction. The > receive queue is fairly full and a block is sent to the source nodes to let > them know not to send anything until the receive queue has been processed. > This could be the result of a large object such as a blob being transmitted > to the target node. > > From: "FRANK" <yunyaoqu@gmail.com> > > To: ids@iiug.org > > Date: 11/15/2011 04:06 PM > > Subject: ER performance [25429] > > Sent by: ids-bounces@iiug.org > > IDS11.50 FC8, Redhat Linux. Three sites ER update anywhere. > > Apparently, we have one site ER constantly queuing data locally, which can > not replicate data to another site fast enough. > > The output of the cdr list server and onstat -g nif are as following > ( in the site where ER queues data). > > The "BLOCK" in th line, > > 22 g_ncdcops RUN,BLOCK 9 220847207 111623315 > > means, sLow network? slow server?, or .... ? (The other two sites do not > have "BLOCK" there.) > > More comments? > > Thanks, > Frank > > [informix@maggie ~]$ cdr list serve > SERVER ID STATE STATUS QUEUE CONNECTION CHANGED > ----------------------------------------------------------------------- > g_ncdcops 22 Active Connected 710147964 Nov 15 20:06:57 > g_ngdcops 23 Active Connected 219989883 Nov 15 21:24:32 > g_nsofops 21 Active Local 0 > > [informix@maggie ~]$ onstat -g nif > IBM Informix Dynamic Server Version 11.50.FC8 -- On-Line -- Up 26 days > 05:41:21 -- 4966528 Kbytes> NIF anchor Block: 1131a77d0 > > nifGState RUN > > RetryTimeout 300 > CDR connections: > Id Name State Version Sent Received > --------------------------------------------------------------------------- > > 22 g_ncdcops RUN,BLOCK 9 220847207 111623315 > 23 g_ngdcops RUN 9 231228996 122563913 > > --0015175cfed4bf019e04b1cd2e4d > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --000e0cdfd67654a4ad04b1ce7569 ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.