john_robinson001 Posted May 19, 2007 Posted May 19, 2007 Hi all, We've been experiencing intermittent issues with performance on our network recently. At seemingly random times in the day, users in the IT suites report their workstations freezing up for minutes on end, documents getting corrupted when they try to save etc. These are all Windows XP machines patches up to SP2. Looking at the server when this is happening, perfmon shows the average disk queue length on the server shooting right off the scale, and sometimes the avg. pages per second too. We're running Server 2003 R2, and the disks are in a RAID-5 configuration. Processor usage stays relatively low. The server in question is a Dual Xeon 3.2GHz with 4Gb RAM. The manufacturer's been out and replaced nearly every component of the server, thinking it was a hardware fault (mainboard, SCSI backplane, CPUs, and nearly all of the disks). None of this has made a blind bit of difference. I'm fast running out of ideas, hence this post!! Any suggestions would be greatly appreciated. Thanks in advance, John Robinson ICT Technician Hawkley Hall High School
wesleyw Posted May 19, 2007 Posted May 19, 2007 Check the event logs what do tthey say? antivirus update? anything else you could tell us may help with the diagnosis. Wes
DMcCoy Posted May 19, 2007 Posted May 19, 2007 Has it got symantec antivirus on it? The only time I've had similar issues were once with Northern Quota server stopping any files being opened on the server, and another time with symantec reacting badly to the way a room full of people working on windows movie maker accessed the server. I guess it could also be a bad disk or disks in the array. I also had issues with a non failed, but broken disk in my xraid box.
john_robinson001 Posted May 20, 2007 Author Posted May 20, 2007 Nothing of particular interest in the event logs, although I'll have another look tomorrow... We run Sophos on our network, so no Symantec to be found, and the server in question isn't our antivirus server. One of our first thoughts was dodgy disks, but the manufacturer has been out and changed most of the disks and the SCSI backplane, to no effect. One thing I have noticed is that this symptom occurs whenever there is a sustained read / write to the disks. E.g. if I run a search on a mapped drive from my PC, the queue length shoots off the scale and the users start ringing up reporting slowdown. Likewise if I copy a few large files to the server. The times that this slowdown occurs seems to be totally intermittent, I've tried disabling DFS to see if folder replication between servers was causing the problem, but that didn't help. I've noticed, however, that this seems to happen on ANY server, even non-RAID servers, if you watch perfmon whilst performing file operations... Maybe its just because this is our main file server we're noticing the issues. Can anyone else report similar? Thanks for the replies so far. John
YouKnowWho Posted May 24, 2007 Posted May 24, 2007 Do you actually have sophos installed on the server? We had a very similar issue it was because of "on-access" scanning within Sophos.
john_robinson001 Posted May 24, 2007 Author Posted May 24, 2007 The server doesn't have the Sophos management console installed, but it does run the client. We had the same thought about on-access scanning and disabled it a while back. It doesn't seem to have made much difference though. Thanks for the reply, John
Geoff Posted May 24, 2007 Posted May 24, 2007 There's a couple of settings that can really bog things down. These are 'Scan Archives' and 'Scan network drives'. Make sure these are disabled.
Quackers Posted May 24, 2007 Posted May 24, 2007 Nothing of particular interest in the event logs, although I'll have another look tomorrow... We run Sophos on our network, so no Symantec to be found, and the server in question isn't our antivirus server. One of our first thoughts was dodgy disks, but the manufacturer has been out and changed most of the disks and the SCSI backplane, to no effect. One thing I have noticed is that this symptom occurs whenever there is a sustained read / write to the disks. E.g. if I run a search on a mapped drive from my PC, the queue length shoots off the scale and the users start ringing up reporting slowdown. Likewise if I copy a few large files to the server. The times that this slowdown occurs seems to be totally intermittent, I've tried disabling DFS to see if folder replication between servers was causing the problem, but that didn't help. I've noticed, however, that this seems to happen on ANY server, even non-RAID servers, if you watch perfmon whilst performing file operations... Maybe its just because this is our main file server we're noticing the issues. Can anyone else report similar? Thanks for the replies so far. John I'm guessing Sophos is the problem, when we had it on workstations and on the 2 we still do the workstation would freeze while it updates, extreamly annoying.
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now