Discussion Locked

Total Cache Buffer diminishing

0 helpful
5 replies
Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.
I was wondering why the "Total Cache Buffers" on my 4.11 Server runs down to almost nothing and then I start to get a message that the system can not allocate memory. I have to down the server and restart. Once this is done, the cache buffers seemto be fine for a while and then it starts again after about a half a day. I have 196MB memory. Orignal Cache buffers on is at about 48,480 and after a reboot the Total Cache Buffers is at about 40,500. Over the last few days, it seems to have gotten worse. I just Rconsoled in from home and with nobody even working on the network, the Total Cache Buffers is still falling off. What happening?? I have other server on the Wan and they seem to be fine

5 replies

by acurrin

A couple things to look at:

Are there any new NLMs running on the server? Anti-virus? GroupWise? CD-ROM software? If so, look into the memory that each NLM is using, note it and then look at each the next day and see if one is climbing.

Also, look at the volume space on all the volumes, if space is low, use of files that are compressed will cause them to be decompressed, but the OS will have to work real hard to manage the files in the minimal free space. Run a purge on the volumes to reduce directory entries that may be eating up RAM and clear up space.

by jaolson40

Also, maybe run a DSREPAIR. I once had a situation exactly like the one you're describing, where I would run out of available memory within a couple of days. Checking the memory stats in MONITOR showed that DS.NLM had the bulk of the RAM, and apparently errors in the NDS database was causing the server to allocate more and more memory to the process. A DSREAPIR fixed it.

Certainly check the memory stats in MONITOR, tho. One glance when memory is low will usually show you the offending NLM orprocess and give you a starting point.

This discussion is closed to new replies.