Add new disc to 3par

[yoshimura1]

hello

Added disk to 3par storage.
The status of the disk added at positions 4:14:0 to 4:19:9 remains new and does not become normal.
------
--- 4:14:0 FC new 1675 0 0 0 0 0 0 0 0 0 0 0
--- 4:15:0 FC new 1675 0 0 0 0 0 0 0 0 0 0 0
--- 4:16:0 FC new 1675 0 0 0 0 0 0 0 0 0 0 0
--- 4:17:0 FC new 1675 0 0 0 0 0 0 0 0 0 0 0
--- 4:18:0 FC new 1675 0 0 0 0 0 0 0 0 0 0 0
--- 4:19:0 FC new 1675 0 0 0 0 0 0 0 0 0 0 0
------

I tried admithw admitpd but no change.
When I ran admithw I got the following output:
------
Admitting PDs: this step can take an extended period of time based on load
0 disks admitted
The following disks are not valid after attempting to admit them:
Id CagePos Type -State- -Detailed_State- -SedState--
--- 4:14:0 FC new normal not_capable
--- 4:15:0 FC new normal not_capable
--- 4:16:0 FC new normal not_capable
--- 4:17:0 FC new normal not_capable
--- 4:18:0 FC new normal not_capable
--- 4:19:0 FC new normal not_capable
------

Any solution?

thank you.

[sbhat09]

Hello @yoshimura1,

Can you please share 'showpd' and 'showpd -i' command outputs for those problematic PDs?

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[yoshimura1]

Thank you for your reply.

Below is the result of the command.
4:14:0 to 4:19:0 are the problematic disks added this time.
We have 6 enclosures, including expansion enclosures, and added 6 1.8T disks each, for a total storage system of 36.
Of these, 6 that were added to Cage 4 have problems.

--------
XXXXXXXX cli% showpd -i
Id CagePos State ----Node_WWN---- --MFR-- -----Model------ -Serial- -FW_Rev- Protocol MediaType -----AdmissionTime-----
--- 4:14:0 new 5000C500E19C61E0 SEAGATE SSKN1800S5xeN010 WBN53XXX 3P04 SAS Magnetic -----------------------
--- 4:15:0 new 5000C500E19C3344 SEAGATE SSKN1800S5xeN010 WBN53XXX 3P04 SAS Magnetic -----------------------
--- 4:16:0 new 5000C500E19C2494 SEAGATE SSKN1800S5xeN010 WBN53XXX 3P04 SAS Magnetic -----------------------
--- 4:17:0 new 5000C500E1A916D8 SEAGATE SSKN1800S5xeN010 WBN55XXX 3P04 SAS Magnetic -----------------------
--- 4:18:0 new 5000C500E19C614C SEAGATE SSKN1800S5xeN010 WBN53XXX 3P04 SAS Magnetic -----------------------
--- 4:19:0 new 5000C500E19C8F50 SEAGATE SSKN1800S5xeN010 WBN53XXX 3P04 SAS Magnetic -----------------------
0 0:0:0 normal 5000C5009FA3AF9C SEAGATE STHB1800S5xeN010 W3Z0JXXX 3P03 SAS Magnetic 2017-08-29 07:03:44 JST
1 0:1:0 normal 5000C5009F53513C SEAGATE STHB1800S5xeN010 W3Z0EXXX 3P03 SAS Magnetic 2017-08-29 07:03:44 JST
2 0:2:0 normal 5000C5009F90556C SEAGATE STHB1800S5xeN010 W3Z0GXXX 3P03 SAS Magnetic 2017-08-29 07:03:44 JST
3 0:3:0 normal 5000C5009FD3D890 SEAGATE STHB1800S5xeN010 W3Z0PXXX 3P03 SAS Magnetic 2017-08-29 07:03:44 JST
4 0:4:0 normal 5000C5009FA3A720 SEAGATE STHB1800S5xeN010 W3Z0JXXX 3P03 SAS Magnetic 2017-08-29 07:03:44 JST

*Omitted hereafter

XXXXXXXX cli% showpd
----Size(MiB)----- ----Ports----
Id CagePos Type RPM State Total Free A B Capacity (GB)
--- 4:14:0 FC 10 new 1715200 0 0:1:1* 1:1:1* 1800
--- 4:15:0 FC 10 new 1715200 0 0:1:1* 1:1:1* 1800
--- 4:16:0 FC 10 new 1715200 0 0:1:1* 1:1:1* 1800
--- 4:17:0 FC 10 new 1715200 0 0:1:1* 1:1:1* 1800
--- 4:18:0 FC 10 new 1715200 0 0:1:1* 1:1:1* 1800
--- 4:19:0 FC 10 new 1715200 0 0:1:1* 1:1:1* 1800
0 0:0:0 FC 10 normal 1715200 122880 0:1:1* 1:1:1* 1800
1 0:1:0 FC 10 normal 1715200 122880 0:1:1* 1:1:1* 1800
2 0:2:0 FC 10 normal 1715200 156672 0:1:1* 1:1:1* 1800
3 0:3:0 FC 10 normal 1715200 141312 0:1:1* 1:1:1* 1800
4 0:4:0 FC 10 normal 1715200 125952 0:1:1* 1:1:1* 1800

*Omitted hereafter
--------

thank you.

[sbhat09]

Hello @yoshimura1 

From the command output, I see that problematic drives are Seagate model 'SSKN1800S5xeN010'.

Have you checked the drive model compatibility with the system?

I think the issue could be because model number SSKN1800S5xeN010 is not among the compatible drives. That is not listed in this list of supported drives - https://support.hpe.com/hpesc/public/docDisplay?docId=emr_na-c04895558

The above document is for 3PAR 8000 series. If your system is not 8000 series, please confirm the model name. I will check the documents for the exact model.

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[yoshimura1]

Thank you for your reply.

Yes, 3par storage and disk compatibility have been confirmed and upgraded to compatible versions in advance.
The minimum OS version in the reference you provided is satisfied.

current version
--------------
Release version 3.3.1 (MU5)
Patches: P126, P132, P135, P140, P146, P150, P151, P155, P156, P170, P172, P173, P175, P177
--------------

The disks added to other enclosures are also using the same model, and they are recognized correctly, so I thought it might be an enclosure problem, but looking at the output of "showcage -d" I don't know the difference from others.

Are there any other points to check?

Thank you.

[sbhat09]

Hello @yoshimura1,

Has HPE support installed the new drives or you have done by yourself?
Can you please provide the detailed procedure that was followed to add the disks?

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[yoshimura1]

Thank you for your reply.

I added the disk myself.
After confirming that there are no errors or hardware abnormalities with 3par in advance, they are inserted into the 6 enclosures in the following order.
The physical arrangement of the enclosures is also the same, inserted in order from the top enclosure.

1st cage 5 slots 14-19
2nd cage 3 slots 14-19
3rd cage 1 slots 14-19
4th.cage0 Slots 14-19
5th cage 2 slots 14-19
6th cage 4 slot 14-19

The disk that is currently NG is the last inserted disk.
From cage5 to cage2, initialization + firmware update ran after insertion, and it was successfully incorporated into CPG.

Thank you.

[sbhat09]

Hello @yoshimura1,

So, the exactly same model of drives you have newly added to the cage 5, 3, 1, 0 and 2 are working fine.
But they are problematic only on cage 4. Correct?

Can you please share 'showcpg -sdg' and 'checkhealth' command outputs?

Also let me know if there are any other alerts.

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[yoshimura1]

Thank you for your reply.

We will notify you of the results of "showcpg -sdg" that have already been obtained.

-----(MiB)------
Id Name Warn Limit Grow Args
3 SAS_RAID6_4D2P - - 65536 -t r6 -ssz 6 -ss 128 -p -devtype FC -rpm 10

I'm not in an environment where I can access 3par right now, so I can't execute the 'checkhealth' command, so I'll contact you again when I get it.

Thank you.

[sbhat09]

Hello @yoshimura1,

Fine. While sending the 'checkhealth' command output, please include 'showcpg -sag' also.

Do you have only one CPG on the system?

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[yoshimura1]

hello

I will inform you the result of "showcpg -sag" once.
-----------
-----(MiB)------
Id Name Warn Limit Grow Args
3 SAS_RAID6_4D2P - - 16384 -p -devtype FC -rpm 10
-----------

I will contact you when I get "checkhealth".

thank you.

[yoshimura1]

hello

I will inform you the result of "showcpg -sag" once.
-----------
-----(MiB)------
Id Name Warn Limit Grow Args
3 SAS_RAID6_4D2P - - 16384 -p -devtype FC -rpm 10
-----------

There is only one CPG, "SAS_RAID6_4D2P".
I will contact you when I get "checkhealth".

thank you.

[sbhat09]

Hello @yoshimura1,

Ok. No problem.

But while waiting for checkhealth output, you can consider this plan as well:
Here the CPG is set with 'mag' level availability. Here while creating an LD, the system has to select 6 (or multiples of 6) chunklets from different drives. Those chunklets can be from the drives in the same cage also.
In this case, there is no problem if there are uneven numbers of drives on the cages.
So, you can distribute those 6 problematic drives on the remaining cages.

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[yoshimura1]

Thank you for your suggestion.

2 node enclosures + 4 expansion enclosures for a total of 6 units are loaded with the same FC disks, and 4 data 2 parity CPGs (SAS_RAID6_4D2P) are configured with HA enclosures, and are designed to withstand enclosure failures.

With HA magazine, the above availability cannot be guaranteed, but we would like to consider it as one of the solutions.

thank you.

[sbhat09]

Hello @yoshimura1,

Are you sure it is HA enclosure or HA cage configuration?
I don't see the 'ha cage' configuration in 'showcpg -sag' and 'showcpg -sdg' outputs.
HA cage configuration shows like this in the 'showcpg -sag' output.

So, please confirm that again. If it is 'ha cage', then please ignore my previous recommendation. That will be good only for 'ha mag' configuration.

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[sbhat09]

Hello @yoshimura1,

Have you confirmed if the configuration is a 'HA mag' or 'HA cage'?

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.