|
|
Re: NFS-write blocks all processes [message #51716 is a reply to message #51707] |
Mon, 20 October 2014 07:11 |
Bernd_K
Messages: 2 Registered: October 2014
|
Junior Member |
|
|
Hi Thorsten.
Thank you very much for your reply.
I am user only, not a sysman, so some of the answers may not be exactly what you are expecting, but I will do my very best
----
- more Info about this setup? - kernel version?
It is a seismic processing environment, 13 nodes dual Xeon E5-2667 v2, Linux version 2.6.18-371.el5 .
----
- any hardware problems?
No hardware problems as such, just a constant fight against runtimes, I/O bottlenecks... Everything is too slow and too small.
This is why the gaps in NFS-read (during NFS-write) really hurt.
My hope was that tweaking some parameters would enable the system to simultaneously read and write from/to disk.
----
- which kind of RAID Controller and Level ist used? - how many Disk / Spindles are used?
22 Disks, 40TB total, RAID controller LS8271-8I, Level 5 and 6 (says sysman)
The above is the "bulk-data" storage, the one which has critical impact on runtime if it does not constantly read/write as much as possible.
There is another RAID of similar size, holding "database" and other stuff, which normally is not constantly active and has no or small throughput during runtime.
Both RAIDs have separate IP-addresses.
----
- post the outpot from nfsstat?
from RAID:
-sh-4.1$ nfsstat -o all
Server packet stats:
packets udp tcp tcpconn
3374939671 494 3374660041 15960
Server rpc stats:
calls badcalls badclnt badauth xdrcall
3375382591 0 0 0 0
Server reply cache:
hits misses nocache
0 91841868 3282784906
Server file handle cache:
lookup anon ncachedir ncachedir stale
0 0 0 0 417
Server nfs v2:
null getattr setattr root lookup readlink
55 11% 178 36% 0 0% 0 0% 15 3% 0 0%
read wrcache write create remove rename
0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
link symlink mkdir rmdir readdir fsstat
0 0% 0 0% 0 0% 0 0% 39 7% 207 41%
Server nfs v3:
null getattr setattr lookup access readlink
205 0% 148779700 4% 128541 0% 120366 0% 1448436 0% 0 0%
read write create mkdir symlink mknod
3082738953 92% 63675228 1% 29584 0% 4483 0% 0 0% 0 0%
remove rmdir rename link readdir readdirplus
24735 0% 3742 0% 166 0% 0 0% 259 0% 238980 0%
fsstat fsinfo pathconf commit
36202899 1% 192 0% 2 0% 10059450 0%
Server nfs v4:
null compound
66 0% 31900709 99%
Server nfs v4 operations:
op0-unused op1-unused op2-future access close commit
0 0% 0 0% 0 0% 325021 0% 257650 0% 321645 0%
create delegpurge delegreturn getattr getfh link
58040 0% 0 0% 3622 0% 29849717 32% 317903 0% 0 0%
lock lockt locku lookup lookup_root nverify
0 0% 0 0% 0 0% 5092 0% 0 0% 0 0%
open openattr open_conf open_dgrd putfh putpubfh
258028 0% 0 0% 74 0% 1 0% 31287868 33% 0 0%
putrootfh read readdir readlink remove rename
128 0% 786775 0% 101225 0% 0 0% 194679 0% 0 0%
renew restorefh savefh secinfo setattr setcltid
612641 0% 0 0% 0 0% 2 0% 967416 1% 65 0%
setcltidconf verify write rellockowner bc_ctl bind_conn
65 0% 0 0% 26753894 29% 0 0% 0 0% 0 0%
exchange_id create_ses destroy_ses free_stateid getdirdeleg getdevinfo
0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
getdevlist layoutcommit layoutget layoutreturn secinfononam sequence
0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
set_ssv test_stateid want_deleg destroy_clid reclaim_comp
0 0% 0 0% 0 0% 0 0% 0 0%
Client packet stats:
packets udp tcp tcpconn
0 0 0 0
Client rpc stats:
calls retrans authrefrsh
565223 89 565223
Client nfs v4:
null read write commit open open_conf
0 0% 953 0% 1 0% 0 0% 2 0% 0 0%
open_noat open_dgrd close setattr fsinfo renew
0 0% 0 0% 2 0% 1 0% 6 0% 564196 99%
setclntid confirm lock lockt locku access
1 0% 1 0% 0 0% 0 0% 0 0% 11 0%
getattr lookup lookup_root remove rename link
12 0% 9 0% 2 0% 0 0% 0 0% 0 0%
symlink create pathconf statfs readlink readdir
0 0% 0 0% 4 0% 8 0% 0 0% 3 0%
server_caps delegreturn getacl setacl fs_locations rel_lkowner
10 0% 0 0% 0 0% 0 0% 0 0% 0 0%
secinfo exchange_id create_ses destroy_ses sequence get_lease_t
0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
reclaim_comp layoutget getdevinfo layoutcommit layoutreturn getdevlist
0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
(null)
0 0%
From one of the nodes:
-bash-3.2$ nfsstat nfs -o all
Server packet stats:
packets udp tcp tcpconn
0 0 0 0
Server rpc stats:
calls badcalls badclnt badauth xdrcall
0 0 0 0 0
Server reply cache:
hits misses nocache
0 0 0
Server file handle cache:
lookup anon ncachedir ncachedir stale
0 0 0 0 0
Client packet stats:
packets udp tcp tcpconn
0 0 0 0
Client rpc stats:
calls retrans authrefrsh
1495170660 33 0
Client nfs v3:
null getattr setattr lookup access readlink
0 0% 228264410 15% 511297 0% 493031 0% 5084816 0% 11463 0%
read write create mkdir symlink mknod
804422516 53% 333380788 22% 81229 0% 8923 0% 260 0% 0 0%
remove rmdir rename link readdir readdirplus
83241 0% 9101 0% 21029 0% 121 0% 3809 0% 313854 0%
fsstat fsinfo pathconf commit
21495660 1% 10 0% 0 0% 100985097 6%
----
Thanks and cheers,
Bernd
|
|
|