Hello there I have a problem with one of our cluster servers that usually hosts exchange on it. There is a defrag process that keeps starting up now and then and consuming up to 100% CPU causing the server to freeze up and exchange to stop responding. The process is dfrgifc.exe and when I checked the process with process explorer the parent was the svchost dcomlaunch process. Whenever we kill this process it seems to start up again but I am unsure what causes it to be started and I'm not too sure on how all this DCOM stuff works. I've checked through scheduled tasks and any similar places but can't find how it starts or what I can do to stop it from running. Anyone know what I can do to find out where it is run from?

Hi Bunta126,You can use process monitor to capture the trace information of when reproducing the issue. Steps: 1. Download process monitor from the link given here http://technet.microsoft.com/en-us/sysinternals/bb896645.aspx 2. Run the process monitor on the problematic server 3. Start Process Monitor. You will see File/Registry activity is being captured. Press Ctrl+E to stop logging data. Press Ctrl+X to clear the currently captured data. Under Process Monitor, press Ctrl+E to start gathering data. 4. Now try to reproduce the issue. Kill the dfrgifc.exe process in task manager, and then wait for it for start up again. 5. Under Process Monitor, Press Ctrl+E to stop capturing. 6. Please save the captured data from Process Monitor to a file (File->Save) and send me this file for further analysis. 7. Meanwhile, please take a screenshot of the task manager.You may send the captured file to us via tfwst@microsoft.com Thanks.This posting is provided "AS IS" with no warranties, and confers no rights.

Thanks, I'll start running that process monitor now, but it might take a while to start up again as it seems to be starting at 2pm on certain days. Will post as soon as I have it captured. EDIT: Nevermind, it just restarted itself briefly soon after I stopped it and then shut down again so I think I have already captured the data you need. There are two process that runs when it starts, the dfrgifc.exe and dfrgntfs.exe but it is only the dfrgifc process that consumes all the CPU. I'll send you the captured log file now.

grrr... fighting with mail server to get zip attachment out atm... Anyway, until the email gets released you can grab a copy of the process monitor logs here: http://dl.dropbox.com/u/37057/LogFile.PML.zip

Hi Bunta126,After researching on the LogFile.PML, I found the first dfrgntfs.exe started up around 1/8/2010 2:37:01 AM, please check the Event Viewer to find what application startup at that time. This can help us to identify which application cause this process startup.This posting is provided "AS IS" with no warranties, and confers no rights.

Going through the event logs it looks like the process is initiated by the Exchange Database Maintenance schedules. However, it seems to be running outside of how we have scheduled them. One is scheduled to run from 5am to 6am and it was started at 7:30am for example. So, if this is what is initiating the defrags, what causes the dfrgifc.exe process to suddenly start using 100% CPU and freezing the server and how do we stop this from happening? The CPU problem seems to happen about once a month or more and the Exchange DB defrags occur daily on each mailbox store.

I'm having the same problem. It's difficult to recreate the event in order to put it into de process monitor and I've been checking the event viewer line per line and I don't see any process that can invoke the dfrgifc.exe. The main problem is that the process takes 100% of the cpu. My case is very similar, it only happens in clustered servers with Windows 2003. I've checked the maintenance plans for SQL and I don't see any related actions that can start the process. I check hidden task in the scheduled task windows, and we don't have tasks there. Any ideas about this process?

Do you have Exchange running on the cluster? Mine seems to be started by the exchange database maintenance tasks but seems to get stuck sometimes and remains running outside of the maintenance schedules. For now we have created some scripts that run a couple times a day checking for the defrag process and kills it if it is running.

Short question : Which backup software you're using? May be I found an article which could be related to this. p.s. hope 6 month later is not to late for a reply :)

Hi, we're using ARC Brighstor 12.5 here. We still have the problem with the processes. Any ideas?

Hi, we're using ARC Brighstor 12.5 here. We still have the problem with the processes. Any ideas? Hi Joswald, For the machines which I support, I figured out that we use ArcServ 12.5 on them too - without sp. I had a look at the ca support site and found that it might be related to arcserv 12.5. There should be a testfix or meanwhile a servicepack available for arcserv 12.5 which should solve the problem. (SP1 I think..) Unfortunately I do not have any feedback from our backup guys, if they already implemented the SP. So therefore - no confirmation on that topic until now. I'll give them a shout tomorrow and see what they come up with. May be you can give it a try too. br, steff

Hi, we're using ARC Brighstor 12.5 here. We still have the problem with the processes. Any ideas? Hi Joswald, For the machines which I support, I figured out that we use ArcServ 12.5 on them too - without sp. I had a look at the ca support site and found that it might be related to arcserv 12.5. There should be a testfix or meanwhile a servicepack available for arcserv 12.5 which should solve the problem. (SP1 I think..) Unfortunately I do not have any feedback from our backup guys, if they already implemented the SP. So therefore - no confirmation on that topic until now. I'll give them a shout tomorrow and see what they come up with. May be you can give it a try too. br, steff Anyone who could confirm this gents?