[Gluster-devel] [client-protocol.c:4605:client_checksum_cbk] no proper reply from server

[Rodney McDuff]

Hi All
   I have 4 sets of 3 deep AFRs unified together with one of the 4 AFRs
as a namespace. (See server and client config attached). The servers and
client are glusterfs-1.3.8pre4 running on Mac OSX 10.5.2.

The client volume mounts ok and I can read and write files to it.
Looking at the server volumes I see files appear in the namespace brick
and storage bricks as expected for the ALU scheduler.

However when I do any fops that fetches fstats (is 'ls -al') I see the
following errors (only from the first brick of each AFR). I don't see
them from subsequent 'ls -al' so I suppose glusterfs is caching these
fstats. When I just do a plain 'ls' (pre-cached) I see no such error.



Apr  2 12:09:26 glusterfs[59032]: E
[client-protocol.c:4605:client_checksum_cbk] ilc1-lectern: no proper
reply from server, returning EINVAL
Apr  2 12:09:26 glusterfs[59032]: E
[client-protocol.c:4605:client_checksum_cbk] ilc1-001: no proper reply
from server, returning EINVAL
Apr  2 12:09:26 glusterfs[59032]: E
[client-protocol.c:4605:client_checksum_cbk] ilc1-003: no proper reply
from server, returning EINVAL
Apr  2 12:09:26 glusterfs[59032]: E
[client-protocol.c:4605:client_checksum_cbk] ilc1-002: no proper reply
from server, returning EINVAL


Also when I try to extract a large tar file (lots of files) in the
gluster filesystem it hangs for a while after just writing a few files.
It needs a remount top become operational again.

-- 
Dr. Rodney G. McDuff                 |Ex ignorantia ad sapientiam
Manager, Strategic Technologies Group|    Ex luce ad tenebras
Information Technology Services      |
The University of Queensland         |
EMAIL: address@hidden          |
TELEPHONE: +61 7 3365 8220           |

------------- namespace server brick
volume brick
  type storage/posix
  option directory /var/tmp/export
end-volume

volume server
  type protocol/server
  subvolumes brick
  option transport-type tcp/server     # For TCP/IP transport
  option client-volume-filename /opt/local/etc/glusterfs/client.vol
  option auth.ip.brick.allow *
end-volume

------- storage server brick

volume brick
  type storage/posix
  option directory /var/tmp/export
end-volume

volume pbrick
  type features/posix-locks
  option mandatory on          # enables mandatory locking on all files
  subvolumes brick
end-volume


volume server
  type protocol/server
  subvolumes pbrick
  option transport-type tcp/server     # For TCP/IP transport
  option client-volume-filename /opt/local/etc/glusterfs/client.vol
  option auth.ip.pbrick.allow *
end-volume




--------- client config






volume ilc1-lectern
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc1-lectern
  option remote-subvolume brick
end-volume

volume ilc3-lectern
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc3-lectern
  option remote-subvolume brick
end-volume

volume ilc4-lectern
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc4-lectern
  option remote-subvolume brick
end-volume

volume AFR000
  type cluster/afr
  subvolumes ilc1-lectern ilc3-lectern ilc4-lectern
end-volume

volume ilc1-001
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc1-001
  option remote-subvolume pbrick
end-volume

volume ilc3-101
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc3-101
  option remote-subvolume pbrick
end-volume

volume ilc4-151
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc4-151
  option remote-subvolume pbrick
end-volume

volume AFR001
  type cluster/afr
  subvolumes ilc1-001 ilc3-101 ilc4-151
end-volume

volume ilc1-002
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc1-002
  option remote-subvolume pbrick
end-volume

volume ilc3-102
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc3-102
  option remote-subvolume pbrick
end-volume

volume ilc4-152
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc4-152
  option remote-subvolume pbrick
end-volume

volume AFR002
  type cluster/afr
  subvolumes ilc1-002 ilc3-102 ilc4-152
end-volume

volume ilc1-003
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc1-003
  option remote-subvolume pbrick
end-volume

volume ilc3-103
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc3-103
  option remote-subvolume pbrick
end-volume

volume ilc4-153
  type protocol/client
  option transport-type tcp/client
  option remote-host ilc4-153
  option remote-subvolume pbrick
end-volume

volume AFR003
  type cluster/afr
  subvolumes ilc1-003 ilc3-103 ilc4-153
end-volume

volume bricks
  type cluster/unify
  option namespace AFR000 # this will not be storage child of unify.
  subvolumes AFR001 AFR002 AFR003
### ** ALU Scheduler Option **
  option scheduler alu
  option alu.limits.min-free-disk  5% #%
  option alu.limits.max-open-files 10000
  option alu.order 
disk-usage:read-usage:write-usage:open-files-usage:disk-speed-usage
  option alu.disk-usage.entry-threshold 2GB
  option alu.disk-usage.exit-threshold  128MB
  option alu.open-files-usage.entry-threshold 1024
  option alu.open-files-usage.exit-threshold 32
  option alu.read-usage.entry-threshold 20 #%
  option alu.read-usage.exit-threshold 4 #%
  option alu.write-usage.entry-threshold 20 #%
  option alu.write-usage.exit-threshold 4 #%
  option alu.disk-speed-usage.entry-threshold 0 # DO NOT SET IT. SPEED IS 
CONSTANT!!!.
  option alu.disk-speed-usage.exit-threshold 0 # DO NOT SET IT. SPEED IS 
CONSTANT!!!.
  option alu.stat-refresh.interval 10sec
  option alu.stat-refresh.num-file-create 10
### ** Random Scheduler **
# option scheduler random
### ** NUFA Scheduler **
# option scheduler nufa
# option nufa.local-volume-name posix1
### ** Round Robin (RR) Scheduler **
# option scheduler rr
# option rr.limits.min-free-disk 5% #%
end-volume