[389-users] Re: disk i/o: very high write rates and poor search performance

2018-09-07 Thread Jan Kowalsky
Hi all, I just wanted to say, that the high i/o rates where gone more or less spontaneously. Probably they where related to multimaster replication and writing data to changelogdb. But we could not trigger down the reason. After one restart the high i/o rates of one of the three suppliers in a

[389-users] Re: disk i/o: very high write rates and poor search performance

2018-08-15 Thread William Brown
On Wed, 2018-08-15 at 11:03 -0600, Rich Megginson wrote: > On 08/15/2018 10:56 AM, David Boreham wrote: > > > > > > On 8/15/2018 10:36 AM, Rich Megginson wrote: > > > > > > Updating the csn generator and the uuid generator will cause a > > > lot of > > > churn in dse.ldif. There are other

[389-users] Re: disk i/o: very high write rates and poor search performance

2018-08-15 Thread Rich Megginson
On 08/15/2018 10:56 AM, David Boreham wrote: On 8/15/2018 10:36 AM, Rich Megginson wrote: Updating the csn generator and the uuid generator will cause a lot of churn in dse.ldif.  There are other housekeeping tasks which will write dse.ldif But if those things were being done so

[389-users] Re: disk i/o: very high write rates and poor search performance

2018-08-15 Thread David Boreham
On 8/15/2018 10:36 AM, Rich Megginson wrote: Updating the csn generator and the uuid generator will cause a lot of churn in dse.ldif.  There are other housekeeping tasks which will write dse.ldif But if those things were being done so frequently that the resulting filesystem I/O showed

[389-users] Re: disk i/o: very high write rates and poor search performance

2018-08-15 Thread Rich Megginson
On 08/15/2018 10:13 AM, David Boreham wrote: in strace.log: [pid 8088] 12:55:39.739539 poll([{fd=435, events=POLLOUT}], 1, 180 [pid 8058] 12:55:39.739573 <... write resumed> ) = 1 <0.87> [pid 8088] 12:55:39.739723 <... poll resumed> ) = 1 ([{fd=435, revents=POLLOUT}]) <0.000168>

[389-users] Re: disk i/o: very high write rates and poor search performance

2018-08-15 Thread David Boreham
in strace.log: [pid 8088] 12:55:39.739539 poll([{fd=435, events=POLLOUT}], 1, 180 [pid 8058] 12:55:39.739573 <... write resumed> ) = 1 <0.87> [pid 8088] 12:55:39.739723 <... poll resumed> ) = 1 ([{fd=435, revents=POLLOUT}]) <0.000168> [pid 8058] 12:55:39.739754 write(426, "dn:

[389-users] Re: disk i/o: very high write rates and poor search performance

2018-08-15 Thread Jan Kowalsky
Hi all, thanks for answering, it took some time until I came again to this problem. Am 09.08.2018 um 10:44 schrieb Ludwig Krispenz: > > On 08/09/2018 01:53 AM, William Brown wrote: >> Sadly this doesn't tell us much :( > we could get a pstack along with iotop to see which threads do teh IO,

[389-users] Re: disk i/o: very high write rates

2018-08-09 Thread David Boreham
On 8/9/2018 2:44 AM, Ludwig Krispenz wrote: Sadly this doesn't tell us much :( we could get a pstack along with iotop to see which threads do teh IO, regular mods or the BDB regulars like trickle, checkpointing Also : strace ___ 389-users

[389-users] Re: disk i/o: very high write rates

2018-08-09 Thread Ludwig Krispenz
On 08/09/2018 01:53 AM, William Brown wrote: In the audit-log there is nothing what would explain this. But in iotop I see a lot of threads like: The audit log itself (and search log) will generate IO themself :) 1621 be/4 dirsrv 0.00 B/s3.95 K/s 0.00 % 0.46 % ns-slapd -D

[389-users] Re: disk i/o: very high write rates

2018-08-08 Thread William Brown
> > In the audit-log there is nothing what would explain this. But in > iotop > I see a lot of threads like: The audit log itself (and search log) will generate IO themself :) > > 1621 be/4 dirsrv 0.00 B/s3.95 K/s 0.00 % 0.46 % ns-slapd > -D > /etc/dirsrv/slapd-ldap0 -i