Bug 909053 - Gluster-swift does not allow operations on multiple volumes concurrently.
Summary: Gluster-swift does not allow operations on multiple volumes concurrently.
Keywords:
Status: CLOSED WORKSFORME
Alias: None
Product: Red Hat Gluster Storage
Classification: Red Hat Storage
Component: gluster-swift
Version: unspecified
Hardware: Unspecified
OS: Unspecified
low
low
Target Milestone: ---
: ---
Assignee: Luis Pabón
QA Contact: Matt Zywusko
URL:
Whiteboard:
Depends On: 924792
Blocks:
TreeView+ depends on / blocked
 
Reported: 2013-02-08 06:22 UTC by Junaid
Modified: 2016-11-08 22:25 UTC (History)
7 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
: 924792 (view as bug list)
Environment:
Last Closed: 2016-02-17 09:45:34 UTC
Embargoed:


Attachments (Terms of Use)

Description Junaid 2013-02-08 06:22:20 UTC
Description of problem:
At any given point of time, users can only perform operations on only one gluster volume. To use the other volume, the ring files have to be recreated.

Version-Release number of selected component (if applicable):
Master

How reproducible:
Always

Steps to Reproduce:
1. Try to do operations on two different accounts at the same time.
2. Operations on one the account's will fail.
  
Actual results:
Operations on one the account's will fail.

Expected results:
Operations on both the account should complete successfully.

Additional info:

Comment 2 Scott Haines 2013-02-13 22:19:01 UTC
Per Feb-13 bug triage meeting, targeting for 2.1.0.

Comment 3 Vijay Bellur 2013-03-07 08:55:49 UTC
CHANGE: http://review.gluster.org/4485 (object-storage: Restoring multi volume support in UFO.) merged in master by Vijay Bellur (vbellur)

Comment 4 Junaid 2013-03-14 08:52:47 UTC
Fixed in upstream.

Comment 5 Alex Wheeler 2013-03-19 12:48:03 UTC
This only partially corrects the problem.  When this patch is applied against 3.4 Alpha 1, and against 3.3.1-11, everything works fine with just one volume.
With a second volume, all works well when just referencing the first volume, but when referencing the second volume, the container-server requests are against the wrong volume.  Here's a snippet of output from my storage1.log file:
Mar 18 22:01:03 r3master account-server 127.0.0.1 - - [18/Mar/2013:22:01:03 +0000] "HEAD /test/0/AUTH_test" 204 - "txe0a73e5b77fa4d6497c20b5377427388" "-" "-" 0.0114 ""
Mar 18 22:01:03 r3master account-server 127.0.0.1 - - [18/Mar/2013:22:01:03 +0000] "PUT /test/0/AUTH_test/test.txt" 201 - "txe0a73e5b77fa4d6497c20b5377427388" "-" "-" 0.0036 ""
Mar 18 22:01:03 r3master container-server 127.0.0.1 - - [18/Mar/2013:22:01:03 +0000] "PUT /admin/0/AUTH_test/test.txt" 202 - "txe0a73e5b77fa4d6497c20b5377427388" "-" "-" 0.0147

As you can see, the container-server PUT is trying to access the admin volume, instead of the test volume.

Comment 8 Luis Pabón 2013-07-12 18:24:56 UTC
Was this entered for RHS2.0 with or without PDQ patches?


Note You need to log in before you can comment on or make changes to this bug.