Thanks for the reply. The symapi logs have no entries for this event. The time stamp on this repeated error doesn't match any log entries. The only activity other than the regular storage I/O is a scripted multi-hop SRDF/A DR process that starts with a clone from the synchronous R2 devices. All the ECC systems supporting the DR replication have 10 gatekeepers in the symgate list. Wierd...
I think your managment host triggred that message. in order to get rid off that message you need to mask 6 cylinder or 2.81 GB device to your managment host or check the mapping masking and zonning for existing gatekeepers. you should have at least 4 gatekeeper if you have 6 or 8 it will be better.
Thermo .. Usually I run my symmetrix and SDM agents on hosts with at least 4 gatekeepers (the strange little devices you suggest to mask) ... But sometimes I still receive the messages .. Will update this thread if I'll solve my problem
It's a gate that our base software (Solution Enabler) uses when talking to the storage. When talking about TCP/IP you need a socket if you want to send/receive data to another host .. When you talk about Symmetrix/DMX you need a gatekeeper (a disk, in a nutshell) to send commands and receive informations from/to the storage.
Well, it turns out that our production W2K3 clusters has plenty of 2.81 GB gatekeepers but they can't use them. In the release note from solutions enabler or enginuity or somewhere I found a tidbit that stated for Windows clusters, the gatekeeper needs to be 8MB in size AND have a signature. Since these very busy clusters were remotely executing the SYMM commands supporting our DR replication, they had to use the busy SAN drives between disk I/O for inband communication to the DMX...
Zonie, Where did you see that mentioned? I have W2K3 clusters on a DMX-4 with each node masked to four 2.81MB (3 cylinders) gatekeepers and have not had any issues talking to the sym. Not saying that we haven't seen the error, but that it has not been an issue.
dynamox
2 Intern
2 Intern
•
20.4K Posts
0
September 25th, 2007 12:00
on unix /var/log/symapi/log
zonie1
41 Posts
0
September 25th, 2007 14:00
xe2sdc
2 Intern
2 Intern
•
2.8K Posts
0
September 25th, 2007 23:00
I think I'll search in our Primus/Powerlink and let you know if I find something usefull ...
I think that the message is triggered from the SDM agent or the symmetrix agent .. still trying to figure out where the message is coming from
Thermo1
9 Posts
0
October 1st, 2007 10:00
xe2sdc
2 Intern
2 Intern
•
2.8K Posts
0
October 2nd, 2007 00:00
ThX !!
jasonj1
50 Posts
0
October 2nd, 2007 15:00
xe2sdc
2 Intern
2 Intern
•
2.8K Posts
0
October 2nd, 2007 16:00
zonie1
41 Posts
0
February 5th, 2008 13:00
AranH1
2.2K Posts
0
February 5th, 2008 14:00
Where did you see that mentioned? I have W2K3 clusters on a DMX-4 with each node masked to four 2.81MB (3 cylinders) gatekeepers and have not had any issues talking to the sym. Not saying that we haven't seen the error, but that it has not been an issue.
Aran
maggie6
47 Posts
0
February 6th, 2008 03:00
Can anyone tell me what it is related to ? any issues on storage or host?
the below servr plslc091 solaris is conneceted to symm araay.
also want to know the wwn 50:01:04:f0:00:77:49:fb-----what is this related to?
Feb 3 00:34:57 plslc091 lpfc: [ID 728700 kern.warning] WARNING: lpfc9:0203:DIe:Nodev timeout on NPort 50:01:04:f0:00:77:49:fb Data: x2011008 x7 x4
Feb 3 00:34:57 plslc091 lpfc: [ID 728700 kern.warning] WARNING: lpfc7:0203:DIe:Nodev timeout on NPort 50:01:04:f0:00:77:49:fb Data: x2011008 x7 x4
Feb 5 00:01:33 plslc091 lpfc: [ID 728700 kern.warning] WARNING: lpfc9:0203:DIe:Nodev timeout on NPort 50:01:04:f0:00:77:4a:df Data: x2011008 x7 x9