This discussion is locked
XServeGuy

Q: XSan 2.1 Strange ACL

I noticed an unusual ACL on several folders in XSAN admin 2.2

The ACL is called FFFF-EEEE-DDDD-CCCC-BBBB-AAAA82005800

Has anyone seen this before?

Xserve, Mac OS X (10.6.2), XSan 2.2

Posted on Feb 12, 2010 9:41 AM

Close

Q: XSan 2.1 Strange ACL

  • All replies
  • Helpful answers

  • by TGABTG,

    TGABTG TGABTG Feb 12, 2010 3:51 PM in response to XServeGuy
    Level 1 (100 points)
    Feb 12, 2010 3:51 PM in response to XServeGuy
    Are you using AD or OD for user authentication?

    Looks like an AD user ID that isn't getting translated. Either your AD server is down or the translation isn't working. I've had issues with both before.
  • by Strontium90,

    Strontium90 Strontium90 Feb 13, 2010 3:58 AM in response to XServeGuy
    Level 5 (4,087 points)
    Servers Enterprise
    Feb 13, 2010 3:58 AM in response to XServeGuy
    Did you recently reboot the SAN? Was there a DNS or reboot of the AD domain boxes? This is a problem of startup order. You fsm and fsmpm processes are starting before DirectoryServices. This means that the OS can not map permission GUIDs to a human readable name. So far the only way to resolve this after a reboot is to unmount the volume from all systems and then mount them. On occasion, I've needed to also restart DirectoryServices to reset the domain bind. If you restart the DS, give it a minute to bind on complex networks. Use dscl to ensure domain browsing is up or id to validate your are reaching the remote domain.

    Hope this helps. I've considered creating a delay for Xsan but have not gotten around to it.
  • by XServeGuy,

    XServeGuy XServeGuy Feb 13, 2010 7:49 PM in response to XServeGuy
    Level 1 (0 points)
    Feb 13, 2010 7:49 PM in response to XServeGuy
    I had a power issue recently where our large APC unit failed to provide continuous power. The APC problem is another story.... I am working with APC to figure that one out.

    Our AD Server, bound OD Server,TS Server, Xsan, RAID, everything went down hard. It was pretty bad. Luckily, all of the servers came back online and were functional. It is probably, that the hard stop could have caused the AD/OD issue or possibly the start up sequence with the fsm processes.

    I think you have both hit on the probable cause and solution. I appreciate the solutions posted by TGABTG and Strontium90. You both were a big help.

    Mant thanks!!!