I have seen, in this forum, several postings reporting similar symptoms but no obvious final findings or resolution.
For the past 2 months or so we have been experiencing this random stop of the Domino service. No entry in event log, nothing out of the ordinary in the log.nsf (plus no correlation with a particular activity on Domino) and the service restarts no problem. The only thing we have noticed is that this seems to occur while users are on-line, ie: during regular business hours.
Has anyone found the real reason for this behaviour and a resolution? Any help would be greatly appreciated.
BTW, the disks on the server are all fully defragmented, and there are no 3rd partie tools running.
I have a PMR open with Lotus but no resolution as of yet.
Hi, We just started experiencing this issue on several of the servers I run (R6.03)
Has anyone found a resolution to this issue?
We started to see a patern in producing the error and I was wondering if anyone who had this problem could also consistently reproduce this.
Basically, we have a local database on a client machine. We then do a Replication → New Replica and enter a server name. If we accept the default location, instead of typing one in or selecting one (just hit enter) the Domino Service stops and needs to be restarted.
Can anyone let me know if you are able to reproduce this, or have noticed any other patterns that cause the error. I am going to place a support call t oLotus, but wanted to see if others were still having the same issue, or had managed to resolve it.
Please post or contact me directly if you are able to reproduce this.
Hi, after months of investigation we were able to find a pattern in our case. Every time Domino stopped it was being accessed by a user who had a name change, ie: I had run Rename on their Person record. I asked all users who had a name change in the last few months to send me their ID files, I recertified them and sent them back. Once they started using the recerified IDs, the problem has ceased.
Just 2 weeks ago I ran a Rename for another user, and within a day the server began to stop again … that is until I asked the user to send me their ID and I recertified. This was confirmation that the source of our issue was that something went wrong with users’ local ID files after the Rename task was run on their Person document.
I am hoping this is something Lotus will fix in future releases (PS: we never experienced such an issue when running R4 !).
Hi, I had a similar problem the day after I Migrated to 6.5.1. I found some useful tips regarding the NSD*.log files. While searching for “FATAL” in the generated NSD*.log files I found that in my case the trend micro virus scanner task npscan.exe was causing the problem. When I started this task manually domino crashed again but this time he came up again nicely due to the server doc setting “automatically restart server after fault/crash”
After installing patch 3 for Trend Micro Scanmail 2.51 npscan.exe worked fine again.
The Domino Service process runs in a very tight loop, reporting to the Windows service manager every 5 seconds that Domino is running normally. The only thing that can break this loop is the termination of nserver.exe, so the fact that the Domino server process (nservice.exe) is terminating is probably the result of another problem, not a problem in and of itself.
Unfortunately at this point I don’t know what else to offer. nserver.exe shouldn’t simply vanish, of course. Check the console.log file to verify that something isn’t causing a normal exit. Check to see whether the other Domino processes are still running or whether they have also terminated. If they are all gone then that indicates a “normal” shutdown unless something is causing everything to terminate.
Thanks Jim.There is no indication of anything causing the exit. In the NSD files, the fatal error occures on the task nserver and is typically coinciding with a lookup to the address book.
Next time this occurs, I will check if the other domino processes are still running.
It is a tricky problem and I am curious how others who had posted a similar issue a few months back resolved it… unfortunately the threads end with no solution indicated.
Your original post is misleading, but your last post tells the story. nserver.exe is crashing, NSD is running and terminating everything. You do not have Fault Recovery enabled, so the NT service simply terminates, as it should. Get the NSD log to your tech support person and resolve the cause of the crash.
Subject: References, please ?? Domino service is NOT stopping at random
Hi Jim, Subsequent to reading your response, I did some more reading on Fault Recovery (including your LDD articule) and would like to explore enabling it on our domino server. Since you headed the whole effort leading to the roll out of this feature for NT, I was wondering if you can point me to an articule(s) that would best discribe how to set it up.
I am the only Notes admin and developer at our company (in an ever downsized IT dept) and it is our main production server that is experiencing the shutdown/crash issues. With the resulting restraints/demands on my time, any help you can provide in terms of references and good practices advice would greatly appreciated.
Subject: RE: References, please ?? Domino service is NOT stopping at random
You already read the LDD article, I’m not sure how much more there is in terms of elaborating on the feature. It’s very easy to set up, just go to the Basics tab of the server document in the address book. There’s a section called Fault Recovery, just check the enabled box. Do not put anything in the CleanupScript field. If you want to receive an email notification every time the server crashes then add yourself to the notification list. If the change is made remotely then remember to replicate the change. Recycle the server and you’re done.
Subject: RE: References, please ?? Domino service is NOT stopping at random
Thanks Jim, I have enabled Fault recovery and we are planning to update to 6.0.3 over the weekend.
A couple of questions:
(1) We had the “Lotus Domino Server” Service set to restart after a failure (under Recovery tab), however it never worked. It this in any way tied in with the Fault Recovery feature in the Server doc?? ie: could fault recovery being disabled stop the service from restarting automatically when set to do so??
(2) Does transaction logging need to be enabled for Fault Recovery to work ??
Subject: RE: References, please ?? Domino service is NOT stopping at random
(1) We had the “Lotus Domino Server” Service set to restart after a failure (under Recovery tab), however it never worked. It this in any way tied in with the Fault Recovery feature in the Server doc?? ie: could fault recovery being disabled stop the service from restarting automatically when set to do so??
I don’t know why automatically restarting the service didn’t work, but it’s not related to Fault Recovery which works whether or not you’re running as a service (they are completely independent restart mechanisms). When Fault Recovery is enabled the Domino service (nservice.exe) detects the server’s abnormal exit and restarts it. The service itself never exits, so there’s no involment with the Windows Service Manager’s auto-restart mechanism.
(2) Does transaction logging need to be enabled for Fault Recovery to work ??
No, nothing is “required” for Fault Recovery to work, except enabling it of course. But transaction logging plus Fault Recovery is a very good combination because transaction logging will make the recovery of the databases very fast when restarting after a crash, so your server becomes available that much sooner.
Subject: RE: Domino service is NOT stopping at random
I have sent several NSD files to the tech support person and have as a result ran many of the address book related maintenance tasks over and over again. Even created a brand new replica to replace the existing one… but the problem persists. Domino suddenly stopped 3 times this past Monday, but has been stable ever since. That is the issue, the problem is very sporadic. The only thing I did end of day Monday was run: “Tell Collector Quit” … as that was a suggestion in response to a past post someone had submitted who was experiencing the same problem.
I received no responses, obviously, and after weeks of trying to resolved the situation with Lotus Tech support’s and other consultant’s involvement, the issue was never resolved. The customer ended up, tragically, migrating to Exchange! Lotus, are you listening?
I hope they are … I will foward your post to the tech support person who owns my PMR.
I was wondering though, in your experience with other companies, have you encountered such behaviour anywhere else or was it restricted to this one company?
I was thinking of migrating to 6.0.3 or 6.5 in hopes this might resolve the issue. Any thoughts or experience it that respect?
Thanks for the feedback. Much appreciated, sort of nice to know we are not the only ones encountering this strange issue.