On 06/13/2016 04:57 PM, cpol...@surewest.net wrote:
you might want to try the very most recent
kernel in git to see if your problem is fixed.
Is there a specific change that is of interest?
___
CentOS mailing list
CentOS@centos.org
On 2016-06-01 20:07, Kelly Lesperance wrote:
> Software RAID 10. Servers are HP DL380 Gen 8s, with 12x4 TB 7200 RPM drives.
>
> On 2016-06-01, 3:52 PM, "centos-boun...@centos.org on behalf of
> m.r...@5-cent.us"
> wrote:
>
> >Kelly
Software RAID 10. Servers are HP DL380 Gen 8s, with 12x4 TB 7200 RPM drives.
On 2016-06-01, 3:52 PM, "centos-boun...@centos.org on behalf of
m.r...@5-cent.us"
wrote:
>Kelly Lesperance wrote:
>> I did some additional testing - I stopped
Kelly Lesperance wrote:
> I did some additional testing - I stopped Kafka on the host, and kicked
> off a disk check, and it ran at the expected speed overnight. I started
> kafka this morning, and the raid check's speed immediately dropped down to
> ~2000K/Sec.
>
> I then enabled the write-back
I did some additional testing - I stopped Kafka on the host, and kicked off a
disk check, and it ran at the expected speed overnight. I started kafka this
morning, and the raid check's speed immediately dropped down to ~2000K/Sec.
I then enabled the write-back cache on the drives (hdparm -W1
All of our Kafka clusters are fairly write-heavy. The cluster in question is
our second-heaviest – we haven’t yet upgraded the heaviest, due to the issues
we’ve been experiencing in this one.
Here is an iostat example from a host within the same cluster, but without the
RAID check running:
On 05/25/2016 09:54 AM, Kelly Lesperance wrote:
What we're seeing is that when the weekly raid-check script executes,
performance nose dives, and I/O wait skyrockets. The raid check starts out
fairly fast (2K/sec - the limit that's been set), but then quickly drops
down to about
Hi Charles,
Looks to me like all of the drives are performing roughly the same –
there’s certainly not 1 that sticks out (also note this is happening on all 23
nodes in the cluster).
Thanks!
Kelly
[root@r1k1.kafka.log10.blackberry sys] # cat /proc/mdstat
Personalities : [raid10]
On 5/25/2016 12:20 PM, m.r...@5-cent.us wrote:
Also, you could either, on boot, go into the card's firmware interface,
and that'll tell you, somewhere, what the firmware version is. Not sure if
MegaRAID will work with this card - if it does, you really want it..even
though it has an actively
On 2016-05-25 19:13, Kelly Lesperance wrote:
> Hdparm didn’t get far:
>
> [root@r1k1 ~] # hdparm -tT /dev/sda
>
> /dev/sda:
> Timing cached reads: Alarm clock
> [root@r1k1 ~] #
Hi Kelly,
Try running 'iostat -xdmc 1'. Look for a single drive that has
substantially greater await than ~10msec.
Kelly Lesperance wrote:
> Already done – they’re not being very helpful, as we don’t have a support
> contract, just standard warranty.
>
Right. We get support for five years, but then we keep things well past
that, we don't get rid of them till they're dying. (Don't talk to me about
"wasting tax
I should rephrase that – some parts of HP are helping us, but the team I opened
the case with isn’t being very helpful.
On 2016-05-25, 4:29 PM, "Kelly Lesperance" wrote:
>Already done – they’re not being very helpful, as we don’t have a support
>contract, just
Already done – they’re not being very helpful, as we don’t have a support
contract, just standard warranty.
On 2016-05-25, 4:27 PM, "centos-boun...@centos.org on behalf of
m.r...@5-cent.us"
wrote:
>Kelly Lesperance wrote:
>>
Kelly Lesperance wrote:
> LSI/Avago’s web pages don’t have any downloads for the SAS2308, so I think
> I’m out of luck wrt MegaRAID.
>
> Bounced the node, confirmed MPT Firmware 15.10.09.00-IT.
> HP Driver is v 15.10.04.00.
>
> Both are the latest from HP.
>
> Unsure why, but the module itself
LSI/Avago’s web pages don’t have any downloads for the SAS2308, so I think I’m
out of luck wrt MegaRAID.
Bounced the node, confirmed MPT Firmware 15.10.09.00-IT.
HP Driver is v 15.10.04.00.
Both are the latest from HP.
Unsure why, but the module itself reports version 20.100.00.00:
[root@r1k1
I installed the latest firmware and driver (mpt2sas) from HP on one system.
The driver is v20, it appears the firmware may be 15, though:
[ 11.128979] mpt2sas version 20.100.00.00 loaded
[ 11.513836] mpt2sas0: LSISAS2308: FWVersion(15.10.09.00),
ChipRevision(0x05), BiosVersion(07.39.00.00)
John R Pierce wrote:
> On 5/25/2016 11:44 AM, Kelly Lesperance wrote:
>> The HBA is an HP H220.
>
> OH.its a very good idea to verify the driver is at the same revision
> level as the firmware.not 100% sure how you do this under CentOS, my
> H220 system is running FreeBSD, and is at
Hdparm didn’t get far:
[root@r1k1 ~] # hdparm -tT /dev/sda
/dev/sda:
Timing cached reads: Alarm clock
[root@r1k1 ~] #
On 2016-05-25, 2:44 PM, "Kelly Lesperance" wrote:
>The HBA is an HP H220.
>
>We haven’t really benchmarked individual drives – all 12 drives are
On 5/25/2016 11:44 AM, Kelly Lesperance wrote:
The HBA is an HP H220.
OH.its a very good idea to verify the driver is at the same revision
level as the firmware.not 100% sure how you do this under CentOS, my
H220 system is running FreeBSD, and is at revision P20, both firmware
and
On 5/25/2016 11:44 AM, Kelly Lesperance wrote:
The HBA is an HP H220.
for the uninitated, thats a LSI SAS2308, in IT (initiator-terminator) mode.
--
john r pierce, recycling bits in santa cruz
___
CentOS mailing list
CentOS@centos.org
The HBA is an HP H220.
We haven’t really benchmarked individual drives – all 12 drives are utilized in
one RAID-10 array, I’m unsure how we would test individual drives without
breaking the array.
Trying ‘hdparm -tT /dev/sda’ now – it’s been running for 25 minutes so far…
Kelly
On
What is the HBA the drives are attached to?
Have you done a quick benchmark on a single disk to check if this is a
raid problem or further down the stack?
Regards,
Dennis
On 25.05.2016 19:26, Kelly Lesperance wrote:
> [merging]
>
> The HBA the drives are attached to has no configuration that
Kelly Lesperance wrote:
> I’ve posted this on the forums at
> https://www.centos.org/forums/viewtopic.php?f=47=57926=244614#p244614
> - posting to the list in the hopes of getting more eyeballs on it.
>
> We have a cluster of 23 HP DL380p Gen8 hosts running Kafka. Basic specs:
>
> 2x E5-2650
> 128
[merging]
The HBA the drives are attached to has no configuration that I’m aware of. We
would have had to accidentally change 23 of them ☺
Thanks,
Kelly
On 2016-05-25, 1:25 PM, "Kelly Lesperance" wrote:
>They are:
>
>[root@r1k1 ~] # hdparm -I /dev/sda
>
They are:
[root@r1k1 ~] # hdparm -I /dev/sda
/dev/sda:
ATA device, with non-removable media
Model Number: MB4000GCWDC
Serial Number: S1Z06RW9
Firmware Revision: HPGD
Transport: Serial, SATA Rev 3.0
Kelly Lesperance wrote:
> I’ve posted this on the forums at
> https://www.centos.org/forums/viewtopic.php?f=47=57926=244614#p244614
> - posting to the list in the hopes of getting more eyeballs on it.
>
> We have a cluster of 23 HP DL380p Gen8 hosts running Kafka. Basic specs:
>
> 2x E5-2650
> 128
I’ve posted this on the forums at
https://www.centos.org/forums/viewtopic.php?f=47=57926=244614#p244614 -
posting to the list in the hopes of getting more eyeballs on it.
We have a cluster of 23 HP DL380p Gen8 hosts running Kafka. Basic specs:
2x E5-2650
128 GB RAM
12 x 4 TB 7200 RPM SATA
27 matches
Mail list logo