Hey Mike, Thank you for taking the time to reply, I certainly appreciate it!
I was writing a reply answering your questions and running the suggested tests, etc, and came across a config error that was probably the culprit. I had two hosts configured with the same IP, but the target was not raising an error about it and the informational messages were being filtered. Once I enabled those, I saw the mistake. We'll see how the backups go tonight, but I'm betting everything's gonna be fine. Thanks for pointing me in the right direction! -- Paul -----Original Message----- From: Mike Christie [mailto:[email protected]] Sent: Tuesday, December 14, 2010 7:18 PM To: [email protected] Cc: Paul Subject: Re: connection1:0: ping timeout of 5 secs expired, recv timeout 5 / connection1:0: detected conn error (1011) On 12/14/2010 03:12 PM, [email protected] wrote: > Hi all... > > I have four CentOS 5.4 (2.6.18-164.11.1.el5) servers with iscsid > version 2.0-871. Two are misbehaving despite identical configuration. > They all connect to Enhance Tech RS8-IP4 array the same way, directly > NIC-to-NIC without a switch, physically separate from LAN. I created > four targets, one per port, and four separate volumes/LUNs. > > Pasted below is the config and error log. About a minute after a > successful login, the timeouts/errors begin and keep coming constantly > pretty much every minute whenever the session is logged in, regardless > of mount state. The problematic units are also often very slow > logging in, mounting, even directory listing at times. Also, they > sometimes time out and remount the fs read-only in the middle of a > large backup run. > There were some fixes to that code in rhel/centos 5.5 kernel, but I do not think that is what you are hitting. Do you see those ping/nop timeout messages even when you are not doing any IO intensive workload? Did you setup your initiator names (/etc/iscsi/initiatorname.iscsi) or did you let the tools do this? Does each server have a unique initiator name or do some servers have the same value in that file? On the target are there any log messsages? If you set node.conn[0].timeo.noop_out_interval = 0 node.conn[0].timeo.noop_out_timeout = 0 (either set that in iscsid.conf then rerun the discovery command and relogin or run iscsiadm -m node -o update -n node.conn[0].timeo.noop_out_interval -v 0 iscsiadm -m node -o update -n node.conn[0].timeo.noop_out_timeout -v 0 then relogin) this will turn off the iscsi nops/pings. Then if run mkfs and do backups, you should not see the ping timeout messages, but do you see low throughout still? Do you still see "conn error 1011" messages but just missing the ping timeout messages? -- ea926p -- You received this message because you are subscribed to the Google Groups "open-iscsi" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/open-iscsi?hl=en.
