Nothing mentioned in these documents about how to do this!!!
↧
Re: MSA 2060 with SAS add shelf
↧
Re: MSA 2052 Dual Controller
Hello @Staylor76,
That's excellent!
We are extremely glad to know the problem has been resolved and we appreciate you for keeping us posted.
↧
↧
Re: MSA 2060 with SAS add shelf
Cording to the Installation guide, in the "Controller enclosure—rear panel layout", yo can find the next note:
"NOTE Storage enclosures support hot-swap replacement of controller modules, PCMs, disk modules, and I/O modules.
Hot-add of disk enclosures is also supported."
↧
Re: msa2040 - expansion enclosure problems
problem still the same
what i did today on a different msa2040
- shutdown from webinterface
- wait for blue LED
- disconnect power
- power on a new replacement AJ941A without disks
- connect with new sas cables (both controllers to P1 / left port)
- connect power to MSA
- wait , login
see same old problem "some but not all of the PHY lanes ......"
so i have this problems with 3 different MSA , tried 3 pairs of cables , tried 3 different expansion enclosures
maybe a firmware issue ? all msa are on GL225P002-02
added a disk to the empty enclosure , made no difference
↧
COMPONENT NAME: MC Memory Exhaustion
I have an MSA 2060 that goes unstable every day or so. the web interface sometimes won't load, failes login, won't display information etc. eventuall it reports MC Memory Exhaustion and the managment controller reboots it'self. After that it will be fine for a while. Ther are no hardware erroes and the firmware is up to date. Unfortunatly support is usless. I was told because there is no hardware error and the firmware is updated I can pound sand. This is not a solution.
↧
↧
Re: COMPONENT NAME: MC Memory Exhaustion
Hi Michael,
Firmware version IN210R004 has added a new feature to identify management controller over loading and automatic restart to recover from the situation. MC could be considered to be a small Linux server with only around 128MB memory. Frequent polling/sessions from monitoring software like Zabbix/PRTG/Nagios causes MC to move to hung state over a period of time. If you have any monitoring tool to monitor MSA health status, I would suggest reducing the polling frequency to avoid over loading management controller. You could check the sessions from monitoring software under Maintenance > Support > Audit logs.
↧
Re: COMPONENT NAME: MC Memory Exhaustion
Yeah I went down that road. I have the new firmware but it isn't rebooting before things start to get wonky so maybe they should tweak that a little. At least not all the time. I had my security guy exclude them from vulnarability scans but it's still going doing it every couple days. Though now that I look at the log on controller B it's getting ssh spammmed by my Orion server. Not sure why since I'm just using snmp. I wonder if B's IP was used on something else. I'll look into that. Thanks for the reply.
↧
Re: COMPONENT NAME: MC Memory Exhaustion
Hi, I have come across one case where Solarwinds software was causing MC over load. Removing MSA from Solarwinds software monitoring stopped the memory exhaustion alerts. You might need to check with Solarwinds software support team to check for any options to reduce the polling to MSA IP address.
↧
High latency and bad performance
Dear,
We've been experiencing issues on our performance lately.
Our volume consist of 10*ST1200MM0009 disk in a vitrual pool
What must we do to solve this issue?
https://drive.google.com/file/d/1MwM7QTKomKFO9zep4ClGgLV1F8BFIBvU/view?usp=drive_link
https://drive.google.com/file/d/1sgHDWo4UwMAK0iWjVVv9Erx27pk972Nb/view?usp=drive_link
https://drive.google.com/file/d/1_LIZh7Wt1_009R-HHLElJkKptnEm0iPq/view?usp=drive_link
↧
↧
Fault-Tolerant to Straight-through and enpansion enclousure
Hi,
I am installed msa2050 running operation which almost full of stroage capacity. Then, we have a plan to expand SFF enclousure the same as before. Existing msa2050 has 1 main frame and 2 enclosure that used to "Fault-tolerant" cabling. According to SAS cable requirement, we would like to change design "fault-tolerant" to "straight-through" then add new disk SFF enclosure. What would happen during operation time. We could have downtime window from authorized team.
If we change existing msa2050 "fault-tolerant" to "straight-though", the running operation and data could be lose?
I am appreciate single feedback of members,
Regards
Nicholas
↧
Re: High latency and bad performance
Hi,
MSA store log file could be uploaded to the below web link to verify the overall health status and to check for any pending firmware updates:
There were few issues with older version controller firmware in MSA 2060 which triggered latency issues initially.
These firmware issues used to get fixed after the storage controller restart.
If the controller and disk firmware is up-to-date check from Fibre channel switch end (if its present) for any errors in any specific path.
If an HPE support case is logged , HPE support team could also check MSA debug logs for any errors in fibre channel path as well as for any hardware issues.
Disk groups (RAID 5/6) created using power of 2 method would provide significant sequential write performance improvement.
Refer page 13 of MSA best practices guide:
https://www.hpe.com/psnow/doc/a00105260enw.pdf?jumpid=in_pdfviewer-psnow
You may launch an SSH session to MSA controller management IP and execute the below command repeatedly few times when ever you face latency to identify whether the disks are oversubscribed:
show disk-statistics
Abort any scrub jobs before executing the above command.
Consistently high IOPS value of greater than 200 indicate that the disks are oversubscribed as they are 10K rpm hard drives.
↧
How to power off/on HP MSA P2000 G3?
I dusted off my old HP MSA P2000 G3 and HP DL360e, threw in LSI 9200-8e and it works great as my backup machine that turns on during the night to backup everything and then it shuts down. That was the plan, at least.
I need to figure out how do I shut down the P2000 G3? I have AP844A controller inside with is just a raw SAS module, with no interface other than 2xSFF-8088 connector. The unit has no power button nor anything.
I also have one AW592B module which has a management interface but as it is a SmartArray (RAID) module it does not allow me to pass through all the disks for ZFS (or maybe I am wrong and it does?).
↧
Query: How to power off/on HP MSA P2000 G3?
System recommended content:
1. HP P2000 G3 SAS MSA System User Guide (614561-002, July 2011)
2. HP P2000 G3 MSA System CLI Reference Guide
Please click on "Thumbs Up/Kudo" icon to give a "Kudo".
Thank you for being a HPE valuable community member.
↧
↧
Re: Query: How to power off/on HP MSA P2000 G3?
Tnaks for the documents but unfortinatley that can't work.
In document HP P2000 G3 SAS MSA System User Guide it states to use SMU or CLI to shut down controlers. But I have 2xAP844A inside P2000 (AP843A) which does not have any network nor serial connectivity therefore there is no way to connect to it and issue shutdown commands.
↧
Re: Fault-Tolerant to Straight-through and enpansion enclousure
Hi There,
Thank you for reaching out. You can perform this online - https://www.hpe.com/psnow/doc/a00015961enw
Ideally there would be no data loss however, it is still recommended to take a valid data backup and prepare a downtime for this to be implemented.
If you face any issues, please log a support case with us.
↧
Re: Query: How to power off/on HP MSA P2000 G3?
And how do you configure your Volumes?
Both P2000 Controllers have an Ethernet Management Port (4) with an IP Address for configuration. (GUI and Comand Line)
And I assume your Server have also an Ethernet Port.
So send an SSH Shutdown Command by Putty to one of the P2000 Controllers by IP.
If you have really no Ethernet Cable attached to the P2000, there is no other way (by SAS as you expect).
And addition, there is also a Serial Port (3) on both Controllers, see Picture (3) & (4) .
Cali
↧
Re: Fault-Tolerant to Straight-through and enpansion enclousure
Hi Nicholas,
Please schedule a downtime , shutdown storage controllers , allow 3-5 minutes for controllers to shut down, remove power to controller enclosure and attached disk enclosures.
Change the cabling to straight through and attach new disk enclosure.
Provide power to disk enclosures first, allow 3-5 minutes for disks to spin up and then provide power to controller enclosure.
The Pools and volumes should still be accessible / intact after changing the cabling to straight through. Data back up is recommended before performing this activity.
↧
↧
Re: Query: How to power off/on HP MSA P2000 G3?
@John1155
The AW592B is the P2000 G3 SAS Controller. This is the controller for the P2000. This is where you would set an IP address and manage the P2000 (Create VDISKs, Create volumes, configure notifications, manage the controllers, ...) The AP844B is just a 'dumb' expansion IOM with no management.
FYI - all equipment is well past End of Support Life so you are on your own for support/repair/etc ...
It sounds like you direct connected the LSI9200 to the expansion IOM. This will work as it's just a SAS expander and will let the LSI HBA see and manage the drives. You could then use software RAID to create your logical storage. You would not have any control over IOM shutdowns, you would not have any notifications of hardware faults from the hardware (ie. power supply fault)
You could use the P2000 SAS Controller and create a hardware RAID VDISK and then carve volumes and present them as LUNs to the host. You would then have hardware fault notifications and be able to shutdown controllers.
I think what you are looking for is a software way to power down the power supplies and unfortunately there is no internal way to power off the MSA. You could shutdown both controllers to minimize the power draw but the power supplies will still be drawing power and likely spinning the disks. If the P2000 controller is running it can spin down the drives. But at the very minimum the system will have the PSUs running and the fans running.
Your other option would be to have an external power switch that could be software controlled. If you go that way you will want to use the P2000 SAS controller as you can shutdown the controller first then power down. The data is completely written to disk then.
My experience with old hardware is that if it is running it tends to stay running, power cycling tends to bring out the bad behavior (taking longer to power up and become ready, spinning up at all for disks)
↧
MSA 2060 iSCSI Virtual pool overcommitment
Hi,
In March we have purchased two new MSA 2060 iSCSI SANs. They have been configured to run about 70% of their capacity as backup storage for Veeam. Yesterday I went to check on it and I am seeing an error on the health page stating "A virtual pool exceeded its threshold for allocated pages, and the virtual pool is overcommitted." (Unfoaruntely the notifications were configured incorrectly and I wasn't aware of this earlier)
I then went to check the actual volume on the veeam server and I cannot access it at all, the disk is showing as raw. I cannot access the data.
On the SAN itself, it's reporting as healthy- the pool is 31.1TB with 4.795TGB spare. I had the pool overcommit enabled (now I can see this was a mistake). But the actual overcommit is 0 so I am baffled by what is happening?
I have stopped backing up to the other SAN and I'm offloading the data to a spare NAS we have just in case. This SAN is showing the same error but it still works and the data is accessible. Can someone suggest the steps to perhaps rectify this?
I have tried the following things:
-Rebooting the MC
-Rebooting both SC
-Power cycling the unit
-Disconnecting the volume from the host
I would appreciate any insignt.
Thank you
EDIT: It's suggesting to expand the storage, so I will do that on the functining SAN but I really need to get the other one working.
↧
Upgrade SMU from V3 to V4
Hi,
I had a look through the forums but I couldn't find anything. Is it possible to upgrade a MSA 2050 SMU v3 to the latest version v4?
Thank you
↧