Project

General

Profile

Bug #17819

iSCSI LUN ID's all set to 0

Added by survive - over 4 years ago. Updated over 4 years ago.

Status:
Resolved
Priority:
Critical
Assignee:
Vaibhav Chauhan
Category:
OS
Target version:
9.10.1-U2
Seen in:
9.10.1-U1
Severity:
New
Reason for Closing:
Reason for Blocked:
Needs QA:
Yes
Needs Doc:
Yes
Needs Merging:
Yes
Needs Automation:
No
Support Suite Ticket:
n/a
Hardware Configuration:
ChangeLog Required:
No

Description

Hi guys,

I just updated to FreeNAS-9.10.1-U1.

I had one problem that took me a little while to figure out.

I have 2 FreeNAS boxes. I use iSCSI on both of them to share some disk to my ESXi cluster. One box has one LUN shared out, the other has 2. The single LUN box updated just fine, but the box with 2 LUNs gave my ESXi boxes a lot of trouble trying to bring the datastores back online. ESXi could see one of the LUNs on one server but could not access it.

Turns out the LUN ID's for both extents was set to "0", naturally that would give a system fits! I set the second LUN id to 1, re-scanned and got my datastores back.

Might be a total one off event, or it might be a real bug.

-Will

Associated revisions

Revision 5c845c16 (diff)
Added by William Grzybowski over 4 years ago

fix(gui): add current lunid to the list of used ids Ticket: #17819

Revision b0e6e3ca (diff)
Added by William Grzybowski over 4 years ago

fix(gui): add current lunid to the list of used ids Ticket: #17819 (cherry picked from commit 5c845c160a9a96590476e92508abf6721c21f0a1)

Revision 79ad1f93 (diff)
Added by William Grzybowski over 4 years ago

fix(gui): add current lunid to the list of used ids Ticket: #17819 (cherry picked from commit 5c845c160a9a96590476e92508abf6721c21f0a1)

Revision cc209c69 (diff)
Added by William Grzybowski about 4 years ago

fix(db): sync latest migration frozen models mismatch Ticket: #17819

History

#1 Updated by Bonnie Follweiler over 4 years ago

  • Assignee set to Kris Moore

#2 Avatar?id=14398&size=24x24 Updated by Kris Moore over 4 years ago

  • Assignee changed from Kris Moore to Josh Paetzel

#3 Updated by Josh Paetzel over 4 years ago

Are you using FC or iSCSI? Your forum post makes this unclear.

Can you attach a save debug when the system is in the broken state?

#4 Updated by Josh Paetzel over 4 years ago

  • Status changed from Unscreened to Screened
  • Priority changed from No priority to Important

#5 Updated by survive - over 4 years ago

Hi guys,

I'm using iSCSI.

I have fixed the problem, all I had to do was edit the LUN under "Associated Targets" in the iSCSI config.

I can break it again or roll back & upgrade again. That said, it was an easy fix but a pain to diagnose.

-Will

#6 Updated by Kevin J over 4 years ago

I saw the same on my system. One iSCSI target with two extents, both set to zero. I changed one of them to 1 and all was well, but I was confused for a sec. Sorry, don't have a debug from before I 'fixed'.

#7 Updated by Stuart Espey over 4 years ago

I saw this too.

Two extents, originally set to auto, now set to 0.

#8 Updated by Stuart Espey over 4 years ago

In fact, I don't see an option for "auto" anymore.

#9 Updated by Josh Paetzel over 4 years ago

  • Assignee changed from Josh Paetzel to William Grzybowski
  • Priority changed from Important to Critical
  • Target version set to 9.10.1-U2

Ok, there's a real bug here in the migration.

If you have two LUNs for the same target set to auto the migration sets them both to zero.

William, can you fix this?

#10 Updated by William Grzybowski over 4 years ago

  • Status changed from Screened to Needs Developer Review

#11 Avatar?id=14398&size=24x24 Updated by Kris Moore over 4 years ago

  • Status changed from Needs Developer Review to Reviewed
  • Assignee changed from William Grzybowski to Vaibhav Chauhan

Looks good here!

#12 Updated by Vaibhav Chauhan over 4 years ago

  • Status changed from Reviewed to Ready For Release

#13 Updated by Vaibhav Chauhan over 4 years ago

  • Status changed from Ready For Release to Resolved

Also available in: Atom PDF