24GB RAM and 4 vCPUs has nothing to do with disk I/O – it’s still the same
amount of read/write requests going to the disk.
Did i mention we went from 7200k sata in a mirror to 15 SAS on a raid 10 with 6
disk, and it not like my users hit it with anything other than excel/word/ppt
and an occasional quark or photoshop file
Jean-Paul Natola
From: [email protected]
To: [email protected]
Subject: RE: [NTSysADM] event 129 on 2012 hyper-V guest
Date: Wed, 8 Jan 2014 03:20:14 +0000
24GB RAM and 4 vCPUs has nothing to do with disk I/O – it’s still the same
amount of read/write requests going to the disk.
The issue in the thread was for Windows Server 2012, and if you read through
the whole thread, there was eventually a hotfix that
some users were pointed to. However, it does seem that this as a result of
Microsoft issuing a fix, which indicates a software issue, not a hardware issue.
Lastly, all the various workarounds you mention are not suggested by Microsoft
in the thread – they are things that various users
are suggesting.
Cheers
Ken
From: [email protected] [mailto:[email protected]]
On Behalf Of J- P
Sent: Wednesday, 8 January 2014 1:00 PM
To: [email protected]
Subject: RE: [NTSysADM] event 129 on 2012 hyper-V guest
thanks however, those are all up to date
and from the link you sent this is HUGE issue that's pointing to MS with no
real solution going on over 9 months back-
and for only 40 concurrent users (at most) hitting the a file server with 24GB
ram and 4 Virtual processors i suspect this is not load related-
prior to this new box, I had the same amount of users hitting a server with one
dual core cpu with 8gb of ram and never saw a trace of i/o overload just had to
retire it as it was 8 yrs old-
It seems NUMEROUS users have opened PSS cases and no actual resolution just
suggestions like "update the driver" "turn of dedup" "disable VSS" "move your
drive from virtual SCSI to IDE"- seems like they're thrwoing anything out to
see what sticks.
Jean-Paul Natola
From:
[email protected]
To: [email protected]
Subject: RE: [NTSysADM] event 129 on 2012 hyper-V guest
Date: Wed, 8 Jan 2014 01:48:02 +0000
No – I meant drivers on your host. The issue seems to be I/O load – if the
driver on the host is bad, that could be causing a backlog of I/O requests in
the
guest.
From:
[email protected] [mailto:[email protected]]
On Behalf Of J- P
Sent: Wednesday, 8 January 2014 12:21 PM
To: [email protected]
Subject: RE: [NTSysADM] event 129 on 2012 hyper-V guest
That is my exact setup, just to clarify , when you say update the driver , you
are referring to the SCSI driver on the guest os exhibiting the behavior
correct?,
and where would i get it, it just says "Microsft hyper-V SCSI controller
Its not like going to the the LSI website and punching in your model, version
and OS
