Best unofficial Apache Server developers community
Username
Forgot password?
Sign in with Twitter account
Sign in with Facebook account
List archives

Re: mogile rebalance

Re: mogile rebalance
(120 lines)
smoothly change from one network to multiple network?
(20 lines)
Dec 19, 2011
Dormando
Dormando
 this is the policy I ran ..to test 
   
 mogadm rebalance policy --options="from_devices=dev3,dev4
to_devices=dev2,dev5 limit_type=device limit_by=size limit=10g"

 mogadm rebalance test
 Tested rebalance policy...
 Policy: from_devices=dev3,dev4 to_devices=dev2,dev5 limit_type=device
limit_by=size limit=10g

Don't say "dev2,dev5" say "2,5". it's just the numbers.

 when i try to run the rebalance i get 
 mogadm rebalance start
 fsck_running fsck running; cannot run rebalance at same time

 BUT i never ran fsck in the first place  ???

 and fsck status shows

    Running: Yes (on lfvsfcp10057.dn.net)
      Status: 226998401 / 275341929 (82.44%)
        Time: 20006m (189 fids/s; 4260m remain)
  Check Type: Normal (check policy + files)

Note that Time: here is huge. It looks like you never stopped fsck, or it
just never finished before. Make sure it's stopped and try again?


Reply
Tags: policy optionstype device
Messages in this thread
Re: mogile rebalance
reply Re: mogile rebalance
(78 lines) Dec 19, 2011 12:11
reply Re: mogile rebalance
(35 lines) Dec 19, 2011 12:23
reply Re: mogile rebalance
(62 lines) Dec 19, 2011 13:10
reply Re: mogile rebalance
(28 lines) Dec 19, 2011 13:13
reply Re: mogile rebalance
(155 lines) Dec 21, 2011 09:09
reply Re: mogile rebalance
(48 lines) Dec 21, 2011 11:02
Re: mogile rebalance
December 19, 2011 07:56:21 AM
Hi , after adding the new device into the cluster i believe mogile completed replicating the files to the new device dev5 and now I wanted to run rebalance in order to distribute the file system load equally .. this is how the devices look now…
Mogile Capacity
December 7, 2011 10:47:43 PM
Folks, We're working through a sizing exercise as we build out our capacity, and I was wondering if anyone could provide guidance with respect to how large it would be prudent to allow a single MogileFS cluster to grow, in terms of numbers of…
upgrading mogile from 2.34 to 2.54
November 14, 2011 12:34:56 PM
Hi, Dormando, Hope you doing and thanks a lot for your help so far . So we are finally ready for the upgrade 2.54 on our production servers and I a bit skeptical about the upgrade , I fall under the paranoid category of the upgrade wiki lol The…
Experiences with Mogile
March 2, 2011 10:01:22 PM
Hi, My name is Eddie Dickey and I am the Dev Mgr at the Public Library of Science in San Francisco. We are evaluating Mogile and would like to talk to folks who have run Mogile in production. I would really appreciate any assistance. Stuart - I…
Clean Mogile Installation
August 25, 2011 01:21:32 AM
Hi, I'm new to MogileFS, trying it out for a photo storing application. I would like to know what is the easiest way to clean up an existing MogileFS installation so that it can be set up afresh? Is it sufficient to truncate all tables in MySQL…
mogile backup and recovery
December 21, 2010 11:05:40 PM
Hi, I have a setup where there are 2 servers with Mogile trackers running on them, and they are storing devices too. Besides that, there are 2 storage devices as well. Out of the 2 mogile trackers, only ONE is active at a time. The other one is…
Re: Possible split mogile database...
December 10, 2010 02:07:41 AM
> I inherited a dilapidated mogilefs, where drains aren't working, etc, > etc, etc. In the course of my maintenance I noticed that mogadm lists > the "file_lockerdb_host" as the old db server(we migrated to a new > one, I dont know if…
Possible split mogile database...
December 2, 2010 03:56:17 PM
I inherited a dilapidated mogilefs, where drains aren't working, etc, etc, etc. In the course of my maintenance I noticed that mogadm lists the "file_lockerdb_host" as the old db server(we migrated to a new one, I dont know if mogile did). In my…
one host down in the mogile FS group
September 30, 2011 01:03:27 AM
Hi, Theare are 14 file servers per mogile instance and we have 3 instances. We just marked one host down for a day because there was a os disk corruption. We solved it and marked it alive again. After that, Traffic increased rapidly to…
What are folks doing for monitoring / trending of mogile?
September 29, 2010 01:03:41 PM
Hi Folks, We have some basic monitoring in place but I'm looking to add some more trending of mogile performance and was wondering if other folks have information they're willing to share. My hope is to trend a couple different areas of…
Understanding the directory tree structure which mogile creates on disk
November 25, 2010 02:30:11 AM
Hi In the mogdata directory on my disk, I see that a file is stored as the following: mogdata/dev3/0/191/000/0191000953.fid I want to know how is this structure understandable, and what all things about the file is stored in the tables. I'd…
Neo Rebalance!
September 19, 2010 01:58:50 AM
Yo, I've finally uploaded a working version of the rebalance implementation I've been thinking about for forever. It's inteface is very raw, and I'd like to hear feedback about ways to make it easier to use. I'll be filling in a wiki page with…
Rebalance error
June 15, 2011 10:04:39 AM
Hi, !watch is reporting this while running rebalance: :: [replicate(11617)] Rebalance for DevFID[d=133;f=7675034] (http://10.0.1.103:7500/dev133/0/007/675/0007675034.fid) failed: HTTP delete (due to 'did_rebalance;ret=1') failed: can't connect…
Re: Rebalance error
June 16, 2011 01:43:20 AM
No. Just a few per minute... On 06/16/2011 08:39 AM, dormando wrote: > > On Thu, 16 Jun 2011, Arkadi Colson wrote: > >> Can this be the reason that rebalance does not cleanup the over-replicated >> files? > How many of…
t/30-rebalance.t failure with IO::AIO 4.11
November 29, 2011 02:51:45 PM
t/30-rebalance.t fails on my Debian testing laptop, which has libio-aio-perl 4.11-2+b1. Towards the end of the test, I get a lot of 404 errors from DELETE. The 404 errors shuffle the "ok 48" message into the middle of the line and hides it from…
Rebalance hangs
August 31, 2011 07:11:29 PM
I'm using 2.53 and rebalance is hanging. I'm not exactly sure where to poke to figure out why. Nothing indicitave of a problem is showing up in !watch or syslog. Any suggestions? To reproduce the problem I run the following commands: mogadm…
Rebalance does not work well
October 17, 2011 09:54:37 PM
Hi all,I have raised a issue ,see https://issues.apache.org/jira/browse/AMQ-3544.
Rebalance Stuck
May 23, 2011 01:33:56 PM
We have a mogilefs cluster with a stuck rebalance. It won't stop, I can't change settings, and because of this I can't do a fsck. kel### @mogtracker1prod:~$ mogadm rebalance settings rebal_host = mogtracker2prod …
mogilefs rebalance
May 9, 2011 03:58:35 AM
Dear Listmembers, I have a question about mogilefs rebalance. We're using mogilefs in a virtual machine, and we have to migrate it. The question is, that if the rebalance is too slow for us, can we use rsyncing instead? After rsync, we simply…
Re: Does anyone actively use rebalance?
July 31, 2010 03:33:06 AM
Rebalance has been immensely useful to us. When adding new storage nodes, we are able to mitigate the damage from HD failures, are able to add nodes more sparingly, and just generally balances out load across storage nodes. ( 12 storage nodes, 6…
How can I get Cassandra to automatically rebalance?
April 24, 2010
I am looking at a cassandra cluster, but the administration effort seems to be quite high. Is there any way I can configure Cassandra to rebalance…