Re: [Iscsitarget-devel] Stable vs Trunk

[John J. <jo...@jo...>]

Caspar,

Not sure if its related, but I too see <Unknown LUN> from time to time.
For me it's caused by very high load on the iSCSI server in conjunction with issues on the RAID Controller. (a disk gone offline in the RAID and rebuild is failing due to corrupt sectors etc, causing high latency on the I/O)

Usually it fixes itself, but if not, one of these almost always fixes it for me:

*         Restart IET on the ISCSI server

*         /sbin/services.sh restart on the ESXi host

JJ

From: Caspar Smit [mailto:c....@tr...]
Sent: 15. februar 2013 03:02
To: isc...@li...
Subject: [Iscsitarget-devel] Stable vs Trunk

Hi all,

We have an ESXi 4.1 machine connected to an IET (1.4.20.2 stable) server (debian 6 stable) which has some issues now and then.

Sometimes (can take a week, other times half a year) all VM's running on the iSCSI datastore hang and the only way to resolve is to reboot the ESXi host.

We see logging on the iet server like:

iscsi_trgt: Abort Task (01) issued on tid:1 lun:1 by sid:7320548421730816 (Unknown LUN)

Repeated every few seconds.

Since the stable version of IET is not updated for two and a half years, i was wondering if these kind of issues are resolved in the trunk version?
Can anyone confirm the trunk version is stable enough for everyday use?

Furthermore, is IET still developed and is there any ETA for a new stable release?

Kind regards,
Caspar Smit