[gpfsug-discuss] mass recall from on-prem COS using a policy
IBM Spectrum Scale
scale at us.ibm.com
Thu Jul 21 20:18:52 BST 2022
Adding back Amey in the chain.
Paul please explicitly add Amey in your replies.
Regards, The Spectrum Scale (GPFS) team
------------------------------------------------------------------------------------------------------------------
If you feel that your question can benefit other users of Spectrum Scale
(GPFS), then please post it to the public IBM developerWroks Forum at
https://www.ibm.com/developerworks/community/forums/html/forum?id=11111111-0000-0000-0000-000000000479.
If your query concerns a potential software error in Spectrum Scale (GPFS)
and you have an IBM software maintenance contract please contact
1-800-237-5511 in the United States or your local IBM Service Center in
other countries.
The forum is informally monitored as time permits and should not be used
for priority messages to the Spectrum Scale (GPFS) team.
From: "Paul Ward" <p.ward at nhm.ac.uk>
To: "gpfsug main discussion list" <gpfsug-discuss at gpfsug.org>
Date: 20-07-2022 08.20 PM
Subject: [EXTERNAL] Re: [gpfsug-discuss] mass recall from on-prem COS
using a policy
Sent by: "gpfsug-discuss" <gpfsug-discuss-bounces at gpfsug.org>
I used the template policy, substituting where applicable: #
more /gpfs/nhmfsa/custom/valentine-migrations/policy/iac-workspace_axio-imager_2021.pol /*******************************************************************************
ZjQcmQRYFpfptBannerStart
This Message Is From an External Sender
This message came from outside your organization.
ZjQcmQRYFpfptBannerEnd
I used the template policy, substituting where applicable:
#
more /gpfs/nhmfsa/custom/valentine-migrations/policy/iac-workspace_axio-imager_2021.pol
/*******************************************************************************
* Licensed Materials - Property of IBM
*
* OCO Source Materials
*
* (C) Copyright IBM Corp. 2016-2017 All Rights Reserved
*
* The source code for this program is not published or other-
* wise divested of its trade secrets, irrespective of what has
* been deposited with the U.S. Copyright Office.
*******************************************************************************/
define(
exclude_list,
(
FALSE
OR PATH_NAME LIKE '%/.mcstore/%'
)
)
RULE EXTERNAL POOL 'migrate' EXEC '/usr/lpp/mmfs/bin/mmcloudgateway files'
OPTS '-F' ESCAPE '% -'
/*
RULE EXTERNAL POOL 'mcstore' EXEC '/usr/lpp/mmfs/bin/mmcloudgateway files'
OPTS '-F' ESCAPE '% -'
RULE 'dmrecall1' MIGRATE FROM POOL 'mcstore' TO POOL 'system' WHERE
(
(
PATH_NAME LIKE '/gpfs/test_dir/%'
)
AND NOT (exclude_list)
)
*/
RULE 'migrate_bulk_workspace_axioimager_2021' MIGRATE FROM POOL 'migrate'
TO POOL 'data' WHERE
(
(
PATH_NAME LIKE '
/gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/%
'
)
AND NOT (exclude_list)
)
-----------------------------------
I changed to the folder listed in the path and ran this command, with the
defer option:
# pwd
/gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021
]# mmapplypolicy $(pwd) -f $(pwd)/policy-lists -I defer -L 3
-P /gpfs/nhmfsa/custom/valentine-migrations/policy/iac-workspace_axio-imager_2021.pol
The result of the policy was:
[I] Summary of Rule Applicability and File Choices:
Rule# Hit_Cnt KB_Hit Chosen KB_Chosen
KB_Ill Rule
0 994 0 994 1566373796
1566373796 RULE 'migrate_bulk_workspace_axioimager_2021' MIGRATE FROM
POOL 'migrate' TO POOL 'data' WHERE(.)
[I] Filesystem objects with no applicable rules: 53.
[I] GPFS Policy Decisions and File Choice Totals:
Chose to migrate 1566373796KB: 994 of 994 candidates;
1566373796KB of chosen data is illplaced or illreplicated;
Predicted Data Pool Utilization in KB and %:
Pool_Name KB_Occupied KB_Total Percent_Occupied
data 175459129252 245111980032 71.583253185%
system 0 0 0.000000000%
(no user data)
[I] Because some data is illplaced or illreplicated, predicted pool
utilization may be negative and/or misleading!
[I] 2022-07-20 at 14:22:23.943 Policy execution. 0 files dispatched.
[I] A total of 0 files have been migrated, deleted or processed by an
EXTERNAL EXEC/script;
0 'skipped' files and/or errors.
---------------------------------------------------------------------
Its selected about the right number, as there are :
# find -type f | wc -l
996
--------------------------------------------------------
the head of the deferred output file list:
15:27:41 [root at scale-sk-pn-1 2021]# head policy-lists.recall.migrate
9085980 2146238895 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FRobin
Armstrong 2021%2FThumbs.db
30157518 903677244 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-113_RL_s0c0x0-8849y0-8889.tif
30157519 321389117 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-113_TL.czi
30157520 1970925505 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-113_TL_s0c0x0-8846y0-8892.tif
30157521 1773348368 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-113_TL_s0c0x0-8849y0-8889.tif
30157522 2126423502 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-29_RL.czi
30157523 1701851598 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-29_RL_s0c0x0-6359y0-6656.tif
30157524 1844809347 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-29_TL.czi
30157525 912638442 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-29_TL_s0c0x0-6359y0-6656.tif
30157526 1280698844 0 --
%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-29_TL_s0c0x0-6359y0-7609.tif
….
--------------------------
If I try to recall the file from these paths:
# mmcloudgateway files recall
"%2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-29_TL_s0c0x0-6359y0-7609.tif"
mmcloudgateway: Internal cloud services returned an error:
file system object
/gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/%
2Fgpfs%2Fnhmfsa%2Fbulk%2Fshare%2Fworkspace%2Fiac-workspace%2Flight_microscopy%2Fzeiss_axioimager_m2%2F2021%2FMatt
Loader 2021%2FML210215_OT-29_TL_s0c0x0-6359y0-7609.tif does not exist
mmcloudgateway: Command failed. Examine previous error messages to
determine cause.
It seems the highlighted part is the current directory, as I tried from
another location and it picked up the pwd I was in.
-----------------------
If I Give it the correctly formed path, it works:
# mmcloudgateway files recall
"/gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-29_TL_s0c0x0-6359y0-7609.tif"
mmcloudgateway: Command completed.
------------------------------------------
I also ran a policy just to identify migrated files:
define(ismigrated,
( MISC_ATTRIBUTES LIKE '%V%') )
Rule 'migrated' LIST 'migrated_list' WHERE ismigrated
Rule 'resident' LIST 'resident_list' WHERE not (ismigrated)
----------------------------------------------------------------
The head of that file correctly forms the posix path…
15:35:38 [root at scale-sk-pn-1 2021]# head list.migrated_list
9085980 2146238895 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Robin
Armstrong 2021/Thumbs.db
30157518 903677244 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-113_RL_s0c0x0-8849y0-8889.tif
30157519 321389117 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-113_TL.czi
30157520 1970925505 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-113_TL_s0c0x0-8846y0-8892.tif
30157521 1773348368 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-113_TL_s0c0x0-8849y0-8889.tif
30157522 2126423502 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-29_RL.czi
30157523 1701851598 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-29_RL_s0c0x0-6359y0-6656.tif
30157524 1844809347 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-29_TL.czi
30157525 912638442 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-29_TL_s0c0x0-6359y0-6656.tif
30157526 1280698844 0
-- /gpfs/nhmfsa/bulk/share/workspace/iac-workspace/light_microscopy/zeiss_axioimager_m2/2021/Matt
Loader 2021/ML210215_OT-29_TL_s0c0x0-6359y0-7609.tif
What’s going on?
What I have done in the past is take the output from my ‘ismigrated’
policy.
Put recall commands before each path.
Divide the file into 4 parts, and run each part of one of our 4 protocol
nodes.
Manually doing what mmapplypolicy should do!
From: gpfsug-discuss <gpfsug-discuss-bounces at gpfsug.org> On Behalf Of Amey
P Gokhale
Sent: 20 July 2022 07:55
To: scale at us.ibm.com; gpfsug main discussion list
<gpfsug-discuss at gpfsug.org>
Cc: gpfsug-discuss <gpfsug-discuss-bounces at gpfsug.org>
Subject: Re: [gpfsug-discuss] mass recall from on-prem COS using a policy
Paul – Recall policy looks to be correct. I see that it is in line
with /opt/ibm/MCStore/samples/recallFromCloud.template.
At the end of policy run, what does the stats indicate, such as “X
Files migrated, Y skipped/error” etc? I assume the path you have
specified, has the files in ‘non-resident’ state, which you are
trying to recall using the policy. Correct? Reason I ask this is
because you mentioned around 650G data to be recalled, but policy
seems to have chosen ~2.5G data across 4812 files. So if this path is
actively used by users, it is likely that the files may have been
transparently recalled on access, and are in co-resident state
already. Hence no data movement visible.
In /var/MCStore/ras/logs/mcstore.log, do you see any recall specific
errors, when you run the policy?
On our test setup, we are planning to run similar policy once, to see
if error is reproducible. I will share what we find.
AMEY GOKHALE
Senior Software Engineer – Spectrum Scale
Phone: 91-988 100 8675
E-mail: amey.gokhale at in.ibm.com
From: Huzefa H Pancha <hupancha at in.ibm.com> On Behalf Of scale at us.ibm.com
Sent: 19 July 2022 23:42
To: gpfsug main discussion list <gpfsug-discuss at gpfsug.org>; Amey P Gokhale
<amey.gokhale at in.ibm.com>
Cc: gpfsug-discuss <gpfsug-discuss-bounces at gpfsug.org>
Subject: Re: [EXTERNAL] Re: [gpfsug-discuss] mass recall from on-prem COS
using a policy
Hi Amey,
Can you provide them guidance from TCT angle.
Regards, The Spectrum Scale (GPFS) team
------------------------------------------------------------------------------------------------------------------
If you feel that your question can benefit other users of Spectrum Scale
(GPFS), then please post it to the public IBM developerWroks Forum at
https://www.ibm.com/developerworks/community/forums/html/forum?id=11111111-0000-0000-0000-000000000479
.
If your query concerns a potential software error in Spectrum Scale (GPFS)
and you have an IBM software maintenance contract please contact
1-800-237-5511 in the United States or your local IBM Service Center in
other countries.
The forum is informally monitored as time permits and should not be used
for priority messages to the Spectrum Scale (GPFS) team.
Inactive hide details for "Paul Ward" ---19-07-2022 09.34.33 PM---Thank
you. Ill-placed on ESS or COS?"Paul Ward" ---19-07-2022 09.34.33 PM---Thank
you. Ill-placed on ESS or COS?
From: "Paul Ward" <p.ward at nhm.ac.uk>
To: "gpfsug main discussion list" <gpfsug-discuss at gpfsug.org>
Date: 19-07-2022 09.34 PM
Subject: [EXTERNAL] Re: [gpfsug-discuss] mass recall from on-prem COS using
a policy
Sent by: "gpfsug-discuss" <gpfsug-discuss-bounces at gpfsug.org>
Thank you. Ill-placed on ESS or COS? I understood restriping was for NSDs,
so that would be on our ESS not COS? The direction I want to move the files
is from COS to ESS. We do not have AFM enabled, we are using TCT.
ZjQcmQRYFpfptBannerStart
This Message Is From an External Sender
This message came from outside your organization.
ZjQcmQRYFpfptBannerEnd
Thank you.
Ill-placed on ESS or COS?
I understood restriping was for NSDs, so that would be on our ESS not COS?
The direction I want to move the files is from COS to ESS.
We do not have AFM enabled, we are using TCT.
From: gpfsug-discuss <gpfsug-discuss-bounces at gpfsug.org> On Behalf Of IBM
Spectrum Scale
Sent: 18 July 2022 20:35
To: gpfsug main discussion list <gpfsug-discuss at gpfsug.org>; Venkateswara R
Puvvada <vpuvvada at in.ibm.com>
Subject: Re: [gpfsug-discuss] mass recall from on-prem COS using a policy
"KB_Ill" shows how much data are ill placed or ill replicated. They can
be resolved by mmrestripefs or mmrestripefile.
Copying to AFM team regarding recall in AFM-COS environment.
Regards, The Spectrum Scale (GPFS) team
------------------------------------------------------------------------------------------------------------------
If your query concerns a potential software error in Spectrum Scale (GPFS)
and you have an IBM software maintenance contract please contact
1-800-237-5511 in the United States or your local IBM Service Center in
other countries.
Inactive hide details for "Paul Ward" ---07/12/2022 11:40:52 AM---Hi all, I
need to recall from on-prem COS a folder with subfo"Paul Ward"
---07/12/2022 11:40:52 AM---Hi all, I need to recall from on-prem COS a
folder with subfolders and files, approximately 4600 fil
From: "Paul Ward" <p.ward at nhm.ac.uk>
To: "gpfsug-discuss at gpfsug.org" <gpfsug-discuss at gpfsug.org>
Date: 07/12/2022 11:40 AM
Subject: [EXTERNAL] [gpfsug-discuss] mass recall from on-prem COS using a
policy
Sent by: "gpfsug-discuss" <gpfsug-discuss-bounces at gpfsug.org>
Hi all, I need to recall from on-prem COS a folder with subfolders and
files, approximately 4600 files making up 656G. We have a policy that runs
every 30 mins, and I added this line to it:
ZjQcmQRYFpfptBannerStart
This Message Is From an External Sender
This message came from outside your organization.
ZjQcmQRYFpfptBannerEnd
Hi all,
I need to recall from on-prem COS a folder with subfolders and files,
approximately 4600 files making up 656G.
We have a policy that runs every 30 mins, and I added this line to it:
RULE 'migrate_iac-workspace_Chem-labs' MIGRATE FROM POOl 'migrate' TO POOL
'data' WHERE PATH_NAME LIKE
'/gpfs/nhmfsa/bulk/share/workspace/iac-workspace/Chem_Labs/%'
Migrate is an external pool
RULE EXTERNAL POOL 'migrate' EXEC '/usr/lpp/mmfs/bin/mmcloudgateway files'
OPTS '-F' ESCAPE '% -'
And data is the default placement pool
RULE 'Placement' SET POOL 'data'
When it runs it identifies matching files:
Rule# Hit_Cnt KB_Hit Chosen
KB_Chosen KB_Ill Rule
13 4846 2491152 4846
2491152 7056 RULE 'migrate_iac-workspace_Chem-labs'
MIGRATE FROM POOL 'migrate' TO POOL 'data' WHERE(.)
I can’t find what ‘KB_ Ill’ means but nothing is migrating.
Migrations in other policies work, but this is a ‘recall’ policy.
From this document
https://www.ibm.com/docs/en/spectrum-scale/5.1.3?topic=pools-migrate-recall-external
It suggests it should be invoked as a recall.
Any ideas?
Kindest regards,
Paul
Paul Ward
TS Infrastructure Architect
Natural History Museum
T: 02079426450
E: p.ward at nhm.ac.uk
A picture containing drawing
Description automatically generated
_______________________________________________
gpfsug-discuss mailing list
gpfsug-discuss at gpfsug.org
http://gpfsug.org/mailman/listinfo/gpfsug-discuss_gpfsug.org
_______________________________________________
gpfsug-discuss mailing list
gpfsug-discuss at gpfsug.org
http://gpfsug.org/mailman/listinfo/gpfsug-discuss_gpfsug.org
_______________________________________________
gpfsug-discuss mailing list
gpfsug-discuss at gpfsug.org
http://gpfsug.org/mailman/listinfo/gpfsug-discuss_gpfsug.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://gpfsug.org/pipermail/gpfsug-discuss_gpfsug.org/attachments/20220722/c1f18922/attachment-0002.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: graycol.gif
Type: image/gif
Size: 105 bytes
Desc: not available
URL: <http://gpfsug.org/pipermail/gpfsug-discuss_gpfsug.org/attachments/20220722/c1f18922/attachment-0002.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 16769903.jpg
Type: image/jpeg
Size: 5356 bytes
Desc: not available
URL: <http://gpfsug.org/pipermail/gpfsug-discuss_gpfsug.org/attachments/20220722/c1f18922/attachment-0002.jpg>
More information about the gpfsug-discuss
mailing list