[Bugs] [Bug 1190318] New: Write 1000 20M files, then same file's data broken.

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1190318

            Bug ID: 1190318
           Summary: Write 1000 20M files, then same file's data broken.
           Product: GlusterFS
           Version: mainline
         Component: disperse
          Severity: high
          Assignee: bugs at gluster.org
          Reporter: jiademing.dd at gmail.com
                CC: bugs at gluster.org, gluster-bugs at redhat.com



Description of problem:
  I use gluster-master git at 20150112.
I create a disperse volume,Volume Name: test

Type: Disperse
Volume ID: 8b67411b-c6c7-450b-92df-8d036fd52a62
Status: Started
Number of Bricks: 1 x (2 + 1) = 3
Transport-type: tcp
Bricks:
Brick1: node-1:/disk1
Brick2: node-3:/disk2
Brick3: node-2:/disk3
Options Reconfigured:
features.quota: on
performance.high-prio-threads: 64
performance.low-prio-threads: 64
performance.least-prio-threads: 64
performance.normal-prio-threads: 64
performance.io-thread-count: 64
server.allow-insecure: on
features.lock-heal: on
network.ping-timeout: 5
performance.client-io-threads: enable

 I use postmark to write 1000 20M files to mountpoint, then I check every
file's md5, several file's md5 is different from other files. Postmark's conf
as follow:
set size 20480000 20480000 
set number 1000
set subdirectories 10
set location /cluster2/test/hello
set read 65536
set write 65536
set bias create -1
set bias read 0
set transactions 1
run 
quit

But,I write less than 10M 1000 files, then all files' md5 are OK.write 20M 1000
files always have this problem.

Version-Release number of selected component (if applicable):
 release-3.6 and master all have this problem.

How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.