Storage mangement advice
Bob Elzer
bob.elzer at gmail.com
Tue Jan 27 10:04:18 MST 2009
How long after a restart does it become unresponsive ? couple hours ? a day
? a week ?
Any processes running with high usage ? either % cpu, or Time used ?
A ps listing might help us.
Do you run hobbit or Big brother ?
I find running hobbit, lets me see when things change.
-----Original Message-----
From: plug-discuss-bounces at lists.plug.phoenix.az.us
[mailto:plug-discuss-bounces at lists.plug.phoenix.az.us] On Behalf Of Stephen
Sent: Monday, January 26, 2009 8:53 AM
To: Main PLUG discussion list
Subject: Re: Storage mangement advice
ok here is where it gets interesting the system has 0 RX/TX errors and i can
ssh in just fine. its just as responsive as it should be.
And the Volumes are listed and running according to plan:
[root at mim-openfiler ~]# lvm lvdisplay
--- Logical volume ---
LV Name /dev/data/generaldata
VG Name data
LV UUID U38s47-ze5g-d71C-BecT-ILch-l7zx-uGt5Z1
LV Write Access read/write
LV Status available
# open 1
LV Size 500.00 GB
Current LE 16000
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 253:0
--- Logical volume ---
LV Name /dev/data/esxidata
VG Name data
LV UUID 0xH2jc-dJxn-u36f-X7rc-JMed-zBzB-KFipik
LV Write Access read/write
LV Status available
# open 0
LV Size 244.16 GB
Current LE 7813
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 253:1
--- Logical volume ---
LV Name /dev/data/it
VG Name data
LV UUID oG0pCW-KP3L-svit-uIJr-C6Lg-cuV4-hWEJM4
LV Write Access read/write
LV Status available
# open 1
LV Size 500.00 GB
Current LE 16000
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 253:2
--- Logical volume ---
LV Name /dev/data/itresource
VG Name data
LV UUID yTH8FM-WxKT-vOIH-AqnZ-HXzY-i7FG-3DPM7h
LV Write Access read/write
LV Status available
# open 1
LV Size 500.00 GB
Current LE 16000
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 253:3
I want to remove all but the /dev/data/generaldata iSCSI share and create a
new share, however the Web Interface has basically just come to a unuaseable
crawl...
And better yet i have no idea what the web interface is anmed to even try
and restart jsut that daemon.
but having ssh open and running helps, and i can keep looking from there.
On Sun, Jan 25, 2009 at 9:15 PM, Stephen <cryptworks at gmail.com> wrote:
> I will have to see if I left ash open. And I think I used ext3, as for
> the memory available it is at a default so all 8gb.
>
> Ill have to look at the rest.
>
> Thank you.
>
> On 1/25/09, Lisa Kachold <lisakachold at obnosis.com> wrote:
>>
>> Let's dissect this:
>>
>> 1) 3Ware
>> Optimization is key:
>>
>> 2) OS?
>>
>> 3) Openfiler
>>
>> Openfiler is GREAT STUFF. Admin user is sensitive; i.e. you cannot
>> change the admin user name. Also the tree permissions are picky for
>> SMB shares and home folders:
>>
>> So for instance, if this is your tree:
>>
>>
>>
>> VG
>>
>> LV
>>
>> Folder1 <--- allocated as share, cannot be made homes
>> folder
>>
>> Folder2 <--- not allocated as share, can be made homes
>> folder
>>
>>
>> I.E. any folders allocated as shares should not be listed for
>> allocation to "homes"
>>
>> Questions:
>>
>> a) You are not using ZFS are you?
>> b) Are you using tracd?
>> c) What memory do you have allocated to openfiler?
>> d) Openfiler with USB devices and various kernel drivers has known
>> issues with Volume allocation.
>> Solution: Load the volume group and logical volumes manually...
>> fdisk -l shows the partition but lvm vgdisplay says "no volume groups
found"
>> [root at openfiler ~]# lvm vgdisplay
>> No volume groups found
>> [root at openfiler ~]# lvm lvdisplay shows the same message... So the
>> next thing to do is to load the volume group manually..
>> [root at openfiler ~]# lvm vgscan Reading all physical volumes. This
>> may take a while...
>> Found volume group "rac" using metadata type lvm2 [root at openfiler
>> ~]# Now that the volume group is loaded, the lvdisplay shows the
>> logical volumes but they are in "NOT Available" status
>> [root at openfiler ~]# lvm lvdisplay
>> --- Logical volume ---
>> LV Name /dev/rac/crs
>> VG Name rac
>> LV UUID iKo4Jn-s2sX-pVda-RD8n-JxNr-EbiA-azLPIi
>> LV Write Access read/write
>> LV Status NOT available
>> LV Size 50.00 GB
>> Current LE 64
>> Segments 1
>> Allocation inherit
>> Read ahead sectors 0
>> [root at openfiler ~]# Now change the status of the volume manually
>> [root at openfiler ~]# lvm lvchange -a y /dev/rac/crs [root at openfiler
>> ~]# lvm lvdisplay
>> --- Logical volume ---
>> LV Name /dev/rac/crs
>> VG Name rac
>> LV UUID iKo4Jn-s2sX-pVda-RD8n-JxNr-EbiA-azLPIi
>> LV Write Access read/write
>> LV Status available
>> # open 0
>> LV Size 50.00 GB
>> Current LE 64
>> Segments 1
>> Allocation inherit
>> Read ahead sectors 0
>> Block device 253:0
>> [root at openfiler ~]# If there are more logical volumes, manually
>> change the status for each..
>>
>>
>> Long way around, but you could write a script for this, rather than
reboot?
>>
>> www.Obnosis.com | http://wiki.obnosis.com |
>> http://hackfest.obnosis.com | http://nuke.obnosis.com (503)754-4452
>> PLUG HACKFESTS - http://uat.edu Second Saturday of Each Month Noon -
>> 3PM
>>
>>
>>
>>
>>> Date: Sat, 24 Jan 2009 22:35:48 -0700
>>> Subject: Storage mangement advice
>>> From: cryptworks at gmail.com
>>> To: plug-discuss at lists.plug.phoenix.az.us
>>>
>>> I have a storage server that i am running, i gave a while with
>>> openfiler and it has mostly been sucessful, but it is developing a
>>> weird quirk that its really unnaceptable.
>>>
>>> we had created an iSCSI volume and attached it to our server and
>>> have atamoving back and forth, quite happily really. but over time
>>> it has developed this weird sluggisness only on the webadmin side.
>>> the iSCSI performance is fine.. frist tiem we discovered this we had
>>> some downtme available as we needed to reboot hte mainserver so we
>>> rebooted the storage server as well. however the sluggisness is
>>> back... and well this emans the openfiler is holding about 1.7TB of
>>> space hostage and not letting me allocate it. and this is kind of
>>> unacceptable because if i want to admin the server i should not have
>>> to reboot it first.
>>>
>>> the configuration is a Supermicro board 3ware 8006-2 with a pair od
>>> drives with the OS installed, then 6 SATA drives connected as AHCI
>>> fro the data array. useing the 3ware to manage the boot partition
>>> and also Os/swap seemed a logical choice for me. but now im having
>>> second thoughts on useing openfiler but i am at a loss for a plan b.
>>>
>>> the fun part is i have had no luck finding this issue among any
>>> other openfiler users at all. and im faily sure that teh system has
>>> the resources avialable.
>>>
>>> 8gb of ram and a single harpertown 5405 i think is a little overkill
>>> to have "performance issues"
>>>
>>> the hardware was eventually going to be repurposeable to a virtual
>>> host so we werent concerned with "wast" on overkill hardware. but as
>>> we are having to completely reset the network we dont want to buy a
>>> whole lot of hardware that we dont have a clear use in a new home.
>>> or will become "throw away" hardware that will have no eventual
>>> purpose in our new facility.
>>>
>>> --
>>> A mouse trap, placed on top of your alarm clock, will prevent you
>>> from rolling over and going back to sleep after you hit the snooze
button.
>>>
>>> Stephen
>>> ---------------------------------------------------
>>> PLUG-discuss mailing list - PLUG-discuss at lists.plug.phoenix.az.us
>>> To subscribe, unsubscribe, or to change your mail settings:
>>> http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss
>>
>> _________________________________________________________________
>> Windows LiveT: E-mail. Chat. Share. Get more ways to connect.
>> http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t2_allup_explore_012
>> 009
>
> --
> Sent from my mobile device
>
> A mouse trap, placed on top of your alarm clock, will prevent you from
> rolling over and going back to sleep after you hit the snooze button.
>
> Stephen
>
--
A mouse trap, placed on top of your alarm clock, will prevent you from
rolling over and going back to sleep after you hit the snooze button.
Stephen
---------------------------------------------------
PLUG-discuss mailing list - PLUG-discuss at lists.plug.phoenix.az.us
To subscribe, unsubscribe, or to change your mail settings:
http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss
More information about the PLUG-discuss
mailing list