Skip to Main Content

Infrastructure Software

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

How to avoid zpool degraded when ISCSI offline for a few minutes

Andrew WatkinsMar 16 2018 — edited Mar 20 2018

Hello,

Any thoughts how to help me with the following. We have had problems with some network switches which our between our storage system and our servers. So we lose complete network access for a few minutes sometimes. OK! Network needs to be fixed, but in the mean time can I stop my solaris servers going complete offline!

Solaris 11 server mounts some ISCSI disks from the above and when it loses the ISCSI for a couple minutes the ZPOOL go offline (DEGRADED), problem is our Windows servers come back to life whenever (minutes or hours) the network (ISCSI disk) is back online.

I know that is the idea of ZFS but is there anything I can do about it. Timeouts etc. I thought about changing the ISCSI timeouts, but I don't think that would help?

Example output:

Mar 15 21:08:05 hydra iscsi: [ID 644300 kern.notice] NOTICE: iscsi connection(50) unable to connect to target iqn.2016-08.uk.

ac.bbk.dcs:dss.possrv, target address 010.063.011.222

..

Mar 15 21:10:54 hydra DESC: Probe of ZFS device 'id1,sd@x42356d3155314259/a' in pool 'postgressrv_rpool' has failed.

Mar 15 21:10:58 hydra DESC: The number of I/O errors associated with ZFS device 'id1,sd@x42356d3155314259/a' in pool 'postgressrv_rpool' exceeded

acceptable levels.

Mar 15 21:11:03 hydra DESC: A file or directory in pool 'postgressrv_rpool' could not be read due to corrupt data.

Mar 15 21:11:06 hydra iscsi: [ID 732673 kern.info] NOTICE: iscsi session(47) iqn.2016-08.uk.ac.bbk:dss.possrv offline

Mar 15 21:11:06 hydra genunix: [ID 483743 kern.info] /scsi_vhci/disk@g42356d3155314259 (sd7) multipath status: degraded: path 7

iscsi0/disk@0001iqn.2016-08.uk.ac.bbk:dss.possrv0001,0 is offline

Mar 15 21:11:07 hydra DESC: ZFS device 'postgressrv_rpool' in pool 'postgressrv_rpool' has insufficient replicas to continue.

Mar 15 21:11:11 hydra DESC: ZFS device 'id1,sd@x42356d3155314259/a' in pool 'postgressrv_rpool' failed to open.

Mar 15 21:11:15 hydra iscsi: [ID 559844 kern.info] NOTICE: iscsi session(47) iqn.2016-08.uk.ac.bbk:dss.possrv online

Mar 15 21:11:15 hydra genunix: [ID 483743 kern.info] /scsi_vhci/disk@g42356d3155314259 (sd7) multipath status: degraded: path 7

iscsi0/disk@0001iqn.2016-08.uk.ac.bbk:dss.possrv0001,0 is online

Mar 15 21:11:15 hydra iscsi: [ID 559844 kern.info] NOTICE: iscsi session(45) iqn.2016-08.uk.ac.bbk:dss.possrv online

Mar 15 21:11:15 hydra genunix: [ID 530209 kern.info] /scsi_vhci/disk@g42356d3155314259 (sd7) multipath status: optimal: path 15

iscsi0/disk@0000iqn.2016-08.uk.ac.bbk:dss.possrv0001,0 is online: Load balancing: round-robin

Mar 15 21:11:15 hydra DESC: ZFS pool 'postgressrv_rpool' has experienced currently unrecoverable I/O failures.

Thanks,

Andrew

Comments
Locked Post
New comments cannot be posted to this locked post.
Post Details
Locked on Apr 17 2018
Added on Mar 16 2018
4 comments
720 views