VMware Cloud Community
JarryG
Expert
Expert

ESXi 5.1: local datastore "lost" (i/o error, changing capacity, etc.)...

A strange thing happened to me today:

I restarted my ESXi 5.1 server just to find (vSphere Client: Configuration->Storage->Datastores) that one of local datastores is lost. It is a single disk on sata/ahci controller which worked without any problem up to now.

First I tried "Rescan for Datastore" (vSphere Client). Datastore appeared for short time, but with strange properties (931GB, but formated capacity only 464GB). Moreover, when I opened datastore-browser, it was totally empty! I used this one only for local backups so it is not a big loss for me, but anyway. Very strange. Then I clicked on "properties" and tried "refresh", and it dissapeared again. "Rescan for Datastores" did not bring it back this time...

Then I went to ESXi-shell using ssh-client. After logging in I tried "df" but all I got is:

~ # df
Traceback (most recent call last):
  File "/sbin/df", line 99, in <module>
    sys.exit(main(sys.argv))
  File "/sbin/df", line 53, in main
    o = eval(output)
  File "<string>", line 1
    Errors:
          ^
SyntaxError: invalid syntax
~ #

I changed to /vmfs/volumes and tried listing all datastores:

/vmfs/volumes # ls -al
ls: ./507c45d1-578e7fa1-966e-001e67648fd4: Input/output error
drwxr-xr-x    1 root     root           512 Mar 16 16:00 .
drwxr-xr-x    1 root     root           512 Mar 16 15:38 ..
drwxr-xr-x    1 root     root             8 Jan  1  1970 1a0e7f76-91b3179c-515e-c71ac9d4cbbe
drwxr-xr-x    1 root     root             8 Jan  1  1970 507abc97-69483b3e-2baa-001e67648fd4
drwxr-xr-t    1 root     root          1120 Oct 14 13:22 507abc9e-bda3af2d-c9e7-001e67648fd4
drwxr-xr-x    1 root     root             8 Jan  1  1970 507abc9f-8be253c6-bc2b-001e67648fd4
drwxr-xr-t    1 root     root          2380 Feb  2 20:13 507ac185-15295b94-2383-001e67648fd4
lrwxr-xr-x    1 root     root            35 Mar 16 16:00 backup -> 507c45d1-578e7fa1-966e-001e67648fd4
drwxr-xr-x    1 root     root             8 Jan  1  1970 f243f673-bc48c11e-bbb5-622915fcf931
lrwxr-xr-x    1 root     root            35 Mar 16 16:00 guests -> 507ac185-15295b94-2383-001e67648fd4
lrwxr-xr-x    1 root     root            35 Mar 16 16:00 hypervisor -> 507abc9e-bda3af2d-c9e7-001e67648fd4
/vmfs/volumes #

Apparently there is somethig wrong with "507c45d1-578e7fa1-966e-001e67648fd4" to which the link /vmfs/volumes/backup points.

I restarted server again and went to bios this time. That disk is visible, full capacity recognised, and does not report any problems. Booted to esxi, and again the same thing: this local datastore is "invisible". I selected "Rescan" again, and this datastore appeared. This time with capacity 148GB (?), Used 1645140 (no unit), and Free "-1,808,850" (do not know what this means). Refreshed storage, and it dissapeared again.

And whenever it dissapeares, I could add it again with "Add Storage...", because it appeares in the Disk/LUN list, with full and correct capacity 931GB, but no VMFS label (so apparently I'd have to reformat it).

What could all this mean? Is the disk simply dying? BTW it is quite new (bought just 2 months ago).

JarryG

_____________________________________________ If you found my answer useful please do *not* mark it as "correct" or "helpful". It is hard to pretend being noob with all those points! 😉
Reply
0 Kudos
3 Replies
EliGodoy
Contributor
Contributor

I have the same issue in ESXi 3.5, How you solve?

Reply
0 Kudos
KrizProgramm
Contributor
Contributor

Today there is an answer to this?

Reply
0 Kudos
continuum
Immortal
Immortal

The appropriate answer to the post from 2013 would have been:

Not enough input to answer this question.

Ulli

 

 

 


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

Reply
0 Kudos