1 |
On 05/27/2014 02:03 PM, Stefan G. Weichinger wrote: |
2 |
> I think I have some IO-topic going on ... very likely some mismatch of |
3 |
> block sizes ... the hw-raid, then LVM, then the snapshot on top of |
4 |
> that ... and a filesystem with properties as target ... oh my. Chosing |
5 |
> noop as IO-scheduler helps a bit but maybe I have to roll back and |
6 |
> rebuild one of the HW-RAID-Arrays with a different blocksize. Stefan |
7 |
|
8 |
Hi Stefan, |
9 |
block size / stripe size mismatches only really penalise random io, if |
10 |
you are trying to use dd and have slow speeds this would suggest |
11 |
something else is awry. |
12 |
I don't know the c600 rad chip personally, but in trying to google it it |
13 |
appears to be a motherboard based raid device? is it a real raid or |
14 |
fakeraid? |
15 |
|
16 |
I'm a little confused over your setup to help. i'm sorry if there is |
17 |
duplication but it would be useful to have all info in one hit rather |
18 |
than trying to piece it together from all your messages. |
19 |
1. please can you list your hardware raid config. I'm looking for the |
20 |
physical disk sizes, the virtual disks and their raid types. do you have |
21 |
cache enabled on the raid card, is there a background scrub or anything |
22 |
like that running? do you have active seek/ prefetch configured ? |
23 |
parity size being 50% of total size is just odd to me - but i guess |
24 |
these are mirrors ? but it says raid-level3 --- just odd, most setups |
25 |
use raid0(not raid) raid1(mirror) raid5(parity stripe) raid6 (double |
26 |
parity stripe) or combinations, like 50.. raid3 is allocating a single |
27 |
disk to parity but is very rarely used. |
28 |
2. how many other devices are actively doing IO? do you have any other |
29 |
raid cards/io cards of note that might be clashign on the board. |
30 |
3. do you have active I/O when doing your performance tests? if you have |
31 |
several virtual machines running depending on what they are doing they |
32 |
will crucify your access. |
33 |
4. are you using any type of CGroups ? |
34 |
5. i'm also confused over your LVM config. please can you send the |
35 |
output of "vgs" "pvs" and "lvs -a -o +devices" |
36 |
6. please also send the output of "mount" |
37 |
7. do you have atop or iotop that you can use to monitor performance - |
38 |
specifically we are looking for disk ios per device and disk latency per |
39 |
device. both before and during you are trying to run your backup. |
40 |
|
41 |
this should give us a better idea of where the problems lay. |