Sabuj Pattanayek
2014-Jul-16 04:41 UTC
[Samba] smbd's using up 100% of all cpu's and load avg slowly going up
Hi, Running samba sernet 4.1.6-7, I've noticed the load avg slowly / steadily creeping up (.e.g > 100). I'm now noticing that several smbd processes are at 100%. I don't actually notice that much bandwidth usage on the system (e.g. iptraf/iftop). Any idea what's causing this? Restarting smbd helps for a few days, but then the high load avg returns. Thanks, Sabuj
Ray Van Dolson
2014-Jul-16 04:59 UTC
[Samba] smbd's using up 100% of all cpu's and load avg slowly going up
What do you see when you do an strace on one of the offending processes? Sent from my iPad> On Jul 15, 2014, at 9:41 PM, "Sabuj Pattanayek" <sabujp at gmail.com> wrote: > > Hi, > > Running samba sernet 4.1.6-7, I've noticed the load avg slowly / steadily > creeping up (.e.g > 100). I'm now noticing that several smbd processes are > at 100%. I don't actually notice that much bandwidth usage on the system > (e.g. iptraf/iftop). Any idea what's causing this? > > Restarting smbd helps for a few days, but then the high load avg returns. > > Thanks, > Sabuj > -- > To unsubscribe from this list go to the following URL and read the > instructions: https://lists.samba.org/mailman/options/samba >
Giedrius Tuminauskas
2014-Jul-16 11:33 UTC
[Samba] smbd's using up 100% of all cpu's and load avg slowly going up
Hi, I have noticed same behaviour with *samba4-4.0.0.61.rc4.el6* on RedHat EL6.5 * Giedrius Tuminauskas* On 16/07/2014 05:41, Sabuj Pattanayek wrote:> Hi, > > Running samba sernet 4.1.6-7, I've noticed the load avg slowly / steadily > creeping up (.e.g > 100). I'm now noticing that several smbd processes are > at 100%. I don't actually notice that much bandwidth usage on the system > (e.g. iptraf/iftop). Any idea what's causing this? > > Restarting smbd helps for a few days, but then the high load avg returns. > > Thanks, > Sabuj
Maybe Matching Threads
- Infiniband performance issues answered?
- default cluster.stripe-block-size for striped volumes on 3.0.x vs 3.3 beta (128kb), performance change if i reduce to a smaller block size?
- things that break with unix extensions = yes, samba 4.1.5 and osx 10.9 clients?
- 1024 char limit for auth.allow and automatically re-reading auth.allow without having to restart glusterd?
- crash when using the cp command to copy files off a striped gluster dir but not when using rsync