Server 2003 SBS dies randomly
|
From: Angie M Price on 3 May 2010 11:55 I have a client with Server 2003 SBS on a Dell PowerEdge SC440. Every once in a while, the server "stops." This seems to happen randomly. I will describe the latest occasion in detail - previous occurrences exhibit a similar chain of events. Other background details: service pack 2 is installed; this is the only server on the network - Backup Exec, Exchange, everything is all on the one server; I installed this OS in March 2008, and this issue has pretty much plagued the server its whole life. At 12:00:18 AM on 5/2, Backup Exec reported that it couldn't log into the server - "Server does not exist or access denied." (App Log: Source Backup Exec, Event ID 57860) At 12:59:36 AM, Exchange could no longer talk to the domain controller/global catalog server - Process MAD.EXE All Domain Controller Servers in use are not responding (App Log: Source MSExchangeDSAccess, Category Topology, Event ID 2102) & Process MAD.EXE All Global Catalog Servers in use are not responding (App Log: Source MSExchangeDSAccess, Category Topology, Event ID 2103) At 1:00:09 AM, Exchange reported server down - LDAP Bind unsuccessful on directory. Directory returned error: [0x51] Server Down. (App Log: Source MSExchangeAL, Event ID 8026) At 1:01:39 AM, Process STORE.EXE All the DS Servers in domain are not responding. (App Log: Source MSExchangeDSAccess, Event ID 2104) There are scores more errors of this type, but you probably get the idea. They fill the log until the server is shut down via holding in the power button around a quarter to 8 on Monday morning (5/3). The server is completely unresponsive to users and even to logon at the console. One other event of note is registered in the System Log after the server is brought back up. The EventLog states that the previous system shutdown at 2:00:20 PM on 5/2 was unexpected. How it registers that time is beyond me, because the server was actually useless at least 14 hours before that, was not physically turned off until around 7:45 on 5/3, and the messages in the App Log seem to be the same both before and after that time. So my guess is that the problem is not related to Backup Exec or Exchange, but that somehow a vital Active Directory-related service is corrupted and hanging up? But how to fix? Or is there some other explanation? I have worked in network support for over a decade, but have never seen this happen on any other server (Thank God!). I really hope someone here can lend some insight! Angie
From: Al Williams on 3 May 2010 13:11 Have you run the SBS BPA? http://www.sbslinks.com/sbsbpa.htm Also, if you are running file-level antivirus on the server ensure that it is not scanning important system, exchange or database folders as that can cause all sorts of wierd issues: http://www.google.ca/search?sourceid=navclient&ie=UTF-8&rlz=1T4GGLJ_enCA307&q=sbs2003+antivirus+exclusions -- Allan Williams Angie M Price wrote: > I have a client with Server 2003 SBS on a Dell PowerEdge SC440. Every once > in a while, the server "stops." This seems to happen > randomly. I will describe the latest occasion in detail - previous > occurrences exhibit a similar chain of events. > Other background details: service pack 2 is installed; this is the > only server on the network - Backup Exec, Exchange, everything is all > on the one server; I installed this OS in March 2008, and this issue > has pretty much plagued the server its whole life. > At 12:00:18 AM on 5/2, Backup Exec reported that it couldn't log into > the server - "Server does not exist or access denied." (App Log: > Source Backup Exec, Event ID 57860) > At 12:59:36 AM, Exchange could no longer talk to the domain > controller/global catalog server - Process MAD.EXE All Domain > Controller Servers in use are not responding (App Log: Source > MSExchangeDSAccess, Category Topology, Event ID 2102) & Process > MAD.EXE All Global Catalog Servers in use are not responding (App > Log: Source MSExchangeDSAccess, Category Topology, Event ID 2103) > At 1:00:09 AM, Exchange reported server down - LDAP Bind unsuccessful > on directory. Directory returned error: [0x51] Server Down. (App > Log: Source MSExchangeAL, Event ID 8026) > At 1:01:39 AM, Process STORE.EXE All the DS Servers in domain are not > responding. (App Log: Source MSExchangeDSAccess, Event ID 2104) > There are scores more errors of this type, but you probably get the > idea. They fill the log until the server is shut down via holding in > the power button around a quarter to 8 on Monday morning (5/3). The > server is completely unresponsive to users and even to logon at the > console. One other event of note is registered in the System Log > after the server is brought back up. The EventLog states that the > previous system shutdown at 2:00:20 PM on 5/2 was unexpected. How it > registers that time is beyond me, because the server was actually > useless at least 14 hours before that, was not physically turned off > until around 7:45 on 5/3, and the messages in the App Log seem to be > the same both before and after that time. So my guess is that the problem > is not related to Backup Exec or > Exchange, but that somehow a vital Active Directory-related service > is corrupted and hanging up? But how to fix? Or is there some other > explanation? I have worked in network support for over a decade, but have > never > seen this happen on any other server (Thank God!). I really hope > someone here can lend some insight! > Angie
From: Angie M Price on 3 May 2010 14:21 Allan, Thank you for these suggestions! I did exclude some more folders and file extensions from the AV scan, and I ran the SBS BPA. The BPA found only minor issues, and I am fixing those up, but think it unlikely that any of those could cause the issue currently under discussion. Unfortunately, this issue happens so randomly that, even if it isn't fixed, it could be months before it surfaces again. Everyone, Although I am gratefully implementing Allan's suggestions, I think it possible that they won't resolve the problem, and I'm sure other people out there have some more ideas. Are there any additional suggestions on what the root of the problem is and how to fix it? Has anyone ever seen this before? Thanks, Allan and everyone, Angie
From: Al Williams on 3 May 2010 15:44 I'd suspect some service is going offline and taking others with it. Look at the early event log messages and maybe have a look on eventid.net for any information about them. Does it always occur during a BE backup? Any events seen before BE started its backup? Are you current on your BE patches - checked with their support? -- Allan Williams Angie M Price wrote: > Allan, > Thank you for these suggestions! I did exclude some more folders and > file extensions from the AV scan, and I ran the SBS BPA. The BPA > found only minor issues, and I am fixing those up, but think it > unlikely that any of those could cause the issue currently under > discussion. Unfortunately, this issue happens so randomly that, even > if it isn't fixed, it could be months before it surfaces again. > > Everyone, > Although I am gratefully implementing Allan's suggestions, I think it > possible that they won't resolve the problem, and I'm sure other > people out there have some more ideas. Are there any additional > suggestions on what the root of the problem is and how to fix it? Has > anyone ever seen this before? > > Thanks, Allan and everyone, > Angie
From: Jim Behning SBS MVP on 4 May 2010 22:25 On Mon, 3 May 2010 11:55:29 -0400, "Angie M Price" <angiep(a)wf-cpas.com> wrote: > >I have a client with Server 2003 SBS on a Dell PowerEdge SC440. Every once >in a while, the server "stops." This seems to happen randomly. I will >describe the latest occasion in detail - previous occurrences exhibit a >similar chain of events. >Other background details: service pack 2 is installed; this is the only >server on the network - Backup Exec, Exchange, everything is all on the one >server; I installed this OS in March 2008, and this issue has pretty much >plagued the server its whole life. >At 12:00:18 AM on 5/2, Backup Exec reported that it couldn't log into the >server - "Server does not exist or access denied." (App Log: Source Backup >Exec, Event ID 57860) >At 12:59:36 AM, Exchange could no longer talk to the domain >controller/global catalog server - Process MAD.EXE All Domain Controller >Servers in use are not responding (App Log: Source MSExchangeDSAccess, >Category Topology, Event ID 2102) & Process MAD.EXE All Global Catalog >Servers in use are not responding (App Log: Source MSExchangeDSAccess, >Category Topology, Event ID 2103) >At 1:00:09 AM, Exchange reported server down - LDAP Bind unsuccessful on >directory. Directory returned error: [0x51] Server Down. (App Log: Source >MSExchangeAL, Event ID 8026) >At 1:01:39 AM, Process STORE.EXE All the DS Servers in domain are not >responding. (App Log: Source MSExchangeDSAccess, Event ID 2104) >There are scores more errors of this type, but you probably get the idea. >They fill the log until the server is shut down via holding in the power >button around a quarter to 8 on Monday morning (5/3). The server is >completely unresponsive to users and even to logon at the console. One >other event of note is registered in the System Log after the server is >brought back up. The EventLog states that the previous system shutdown at >2:00:20 PM on 5/2 was unexpected. How it registers that time is beyond me, >because the server was actually useless at least 14 hours before that, was >not physically turned off until around 7:45 on 5/3, and the messages in the >App Log seem to be the same both before and after that time. >So my guess is that the problem is not related to Backup Exec or Exchange, >but that somehow a vital Active Directory-related service is corrupted and >hanging up? But how to fix? Or is there some other explanation? >I have worked in network support for over a decade, but have never seen this >happen on any other server (Thank God!). I really hope someone here can >lend some insight! >Angie I had an antivirus program cause a memory leak which caused the server to drop off the network but I could still see it from the console. Uninstall the AV, reboot and reinstall and it worked fine after that. It took a few weeks for tech support to point their finger at the AV. One hopes that the server has plenty of free space on the partitions. See what SBS support is working on http://blogs.technet.com/sbs/default.aspx Check your SBS with the SBS Best Practices Analyzer http://blogs.technet.com/sbs/archive/tags/BPA/default.aspx
|
Pages: 1 Prev: Mac OS X Tiger connect problem with SBS 2008? Next: small business server licensing |