<tt><font size=3>>I ll open a pmr here for my env ... the issue may
hurt you inralf a ces env. only... but needs to be fixed in core gpfs.base
i think</font></tt><br><br><font size=2 face="sans-serif">Thanks for opening the PMR.</font><br><font size=2 face="sans-serif">The problem is inside the gpfs base
code and we are working on a fix right now.</font><br><font size=2 face="sans-serif">In the meantime until the fix is available
we will use the PMR to propose/discuss potential work arounds.</font><br><br><br><font size=2 face="sans-serif">Mit freundlichen Grüßen / Kind regards<br><br>Mathias Dietz<br></font><br><font size=2 face="sans-serif">Spectrum Scale - Release Lead Architect
(4.2.X Release)</font><br><font size=2 face="sans-serif">System Health and Problem Determination
Architect <br>IBM Certified Software Engineer<br><br>----------------------------------------------------------------------------------------------------------<br>IBM Deutschland<br>Hechtsheimer Str. 2<br>55131 Mainz<br>Phone: +49-6131-84-2027<br>Mobile: +49-15152801035<br>E-Mail: mdietz@de.ibm.com<br>----------------------------------------------------------------------------------------------------------<br>IBM Deutschland Research & Development GmbH<br>Vorsitzender des Aufsichtsrats: Martina Koederitz, Geschäftsführung: Dirk
Wittkopp<br>Sitz der Gesellschaft: Böblingen / Registergericht: Amtsgericht Stuttgart,
HRB 243294<br> <br></font><br><br><br><br><font size=1 color=#5f5f5f face="sans-serif">From:
</font><font size=1 face="sans-serif">Olaf Weiser/Germany/IBM@IBMDE</font><br><font size=1 color=#5f5f5f face="sans-serif">To:
</font><font size=1 face="sans-serif">"gpfsug main discussion
list" <gpfsug-discuss@spectrumscale.org></font><br><font size=1 color=#5f5f5f face="sans-serif">Cc:
</font><font size=1 face="sans-serif">"gpfsug main discussion
list" <gpfsug-discuss@spectrumscale.org></font><br><font size=1 color=#5f5f5f face="sans-serif">Date:
</font><font size=1 face="sans-serif">01/31/2017 11:47 PM</font><br><font size=1 color=#5f5f5f face="sans-serif">Subject:
</font><font size=1 face="sans-serif">Re: [gpfsug-discuss]
CES doesn't assign addresses to nodes</font><br><font size=1 color=#5f5f5f face="sans-serif">Sent by:
</font><font size=1 face="sans-serif">gpfsug-discuss-bounces@spectrumscale.org</font><br><hr noshade><br><br><br><font size=3 face="Helvetica">Yeah... depending on the #nodes you 're
affected or not. .....<br>So if your remote ces cluster is small enough in terms of the #nodes
... you'll neuer hit into this issue <br><br>Gesendet von IBM Verse</font><font size=3><br></font><br><font size=3 color=#424282>Simon Thompson (Research Computing - IT
Services) --- Re: [gpfsug-discuss] CES doesn't assign addresses to nodes
--- </font><br><table width=100% style="border-collapse:collapse;"><tr valign=top height=8><td width=8% style="border-style:none none none none;border-color:#000000;border-width:0px 0px 0px 0px;padding:0px 0px;"><font size=2 color=#5f5f5f>Von:</font><td width=91% style="border-style:none none none none;border-color:#000000;border-width:0px 0px 0px 0px;padding:0px 0px;"><font size=2>"Simon
Thompson (Research Computing - IT Services)" <S.J.Thompson@bham.ac.uk></font><tr valign=top height=8><td style="border-style:none none none none;border-color:#000000;border-width:0px 0px 0px 0px;padding:0px 0px;"><font size=2 color=#5f5f5f>An:</font><td style="border-style:none none none none;border-color:#000000;border-width:0px 0px 0px 0px;padding:0px 0px;"><font size=2>"gpfsug
main discussion list" <gpfsug-discuss@spectrumscale.org></font><tr valign=top height=8><td style="border-style:none none none none;border-color:#000000;border-width:0px 0px 0px 0px;padding:0px 0px;"><font size=2 color=#5f5f5f>Datum:</font><td style="border-style:none none none none;border-color:#000000;border-width:0px 0px 0px 0px;padding:0px 0px;"><font size=2>Di.
31.01.2017 21:07</font><tr valign=top height=8><td style="border-style:none none none none;border-color:#000000;border-width:0px 0px 0px 0px;padding:0px 0px;"><font size=2 color=#5f5f5f>Betreff:</font><td style="border-style:none none none none;border-color:#000000;border-width:0px 0px 0px 0px;padding:0px 0px;"><font size=2>Re:
[gpfsug-discuss] CES doesn't assign addresses to nodes</font></table><br><hr noshade><br><tt><font size=3>We use multicluster for our environment, storage systems
in a separate cluster to hpc nodes on a separate cluster from protocol
nodes.<br><br>According to the docs, this isn't supported, but we haven't seen any issues.
Note unsupported as opposed to broken.<br><br>Simon<br>________________________________________<br>From: gpfsug-discuss-bounces@spectrumscale.org [gpfsug-discuss-bounces@spectrumscale.org]
on behalf of Jonathon A Anderson [jonathon.anderson@colorado.edu]<br>Sent: 31 January 2017 17:47<br>To: gpfsug main discussion list<br>Subject: Re: [gpfsug-discuss] CES doesn't assign addresses to nodes<br><br>Yeah, I searched around for places where ` tsctl shownodes up` appears
in the GPFS code I have access to (i.e., the ksh and python stuff); but
it’s only in CES. I suspect there just haven’t been that many people
exporting CES out of an HPC cluster environment.<br><br>~jonathon<br><br><br>From: <gpfsug-discuss-bounces@spectrumscale.org> on behalf of Olaf
Weiser <olaf.weiser@de.ibm.com><br>Reply-To: gpfsug main discussion list <gpfsug-discuss@spectrumscale.org><br>Date: Tuesday, January 31, 2017 at 10:45 AM<br>To: gpfsug main discussion list <gpfsug-discuss@spectrumscale.org><br>Subject: Re: [gpfsug-discuss] CES doesn't assign addresses to nodes<br><br>I ll open a pmr here for my env ... the issue may hurt you in a ces env.
only... but needs to be fixed in core gpfs.base i thi k<br><br>Gesendet von IBM Verse<br>Jonathon A Anderson --- Re: [gpfsug-discuss] CES doesn't assign addresses
to nodes ---<br><br>Von:<br><br>"Jonathon A Anderson" <jonathon.anderson@colorado.edu><br><br>An:<br><br>"gpfsug main discussion list" <gpfsug-discuss@spectrumscale.org><br><br>Datum:<br><br>Di. 31.01.2017 17:32<br><br>Betreff:<br><br>Re: [gpfsug-discuss] CES doesn't assign addresses to nodes<br><br>________________________________<br><br>No, I’m having trouble getting this through DDN support because, while
we have a GPFS server license and GRIDScaler support, apparently we don’t
have “protocol node” support, so they’ve pushed back on supporting this
as an overall CES-rooted effort.<br><br>I do have a DDN case open, though: 78804. If you are (as I suspect) a GPFS
developer, do you mind if I cite your info from here in my DDN case to
get them to open a PMR?<br><br>Thanks.<br><br>~jonathon<br><br><br>From: <gpfsug-discuss-bounces@spectrumscale.org> on behalf of Olaf
Weiser <olaf.weiser@de.ibm.com><br>Reply-To: gpfsug main discussion list <gpfsug-discuss@spectrumscale.org><br>Date: Tuesday, January 31, 2017 at 8:42 AM<br>To: gpfsug main discussion list <gpfsug-discuss@spectrumscale.org><br>Subject: Re: [gpfsug-discuss] CES doesn't assign addresses to nodes<br><br>ok.. so obviously ... it seems , that we have several issues..<br>the 3983 characters is obviously a defect<br>have you already raised a PMR , if so , can you send me the number ?<br><br><br><br><br>From: Jonathon A Anderson <jonathon.anderson@colorado.edu><br>To: gpfsug main discussion list <gpfsug-discuss@spectrumscale.org><br>Date: 01/31/2017 04:14 PM<br>Subject: Re: [gpfsug-discuss] CES doesn't assign
addresses to nodes<br>Sent by: gpfsug-discuss-bounces@spectrumscale.org<br>________________________________<br><br><br><br>The tail isn’t the issue; that’ my addition, so that I didn’t have to
paste the hundred or so line nodelist into the thread.<br><br>The actual command is<br><br>tsctl shownodes up | $tr ',' '\n' | $sort -o $upnodefile<br><br>But you can see in my tailed output that the last hostname listed is cut-off
halfway through the hostname. Less obvious in the example, but true, is
the fact that it’s only showing the first 120 hosts, when we have 403
nodes in our gpfs cluster.<br><br>[root@sgate2 ~]# tsctl shownodes up | tr ',' '\n' | wc -l<br>120<br><br>[root@sgate2 ~]# mmlscluster | grep '\-opa' | wc -l<br>403<br><br>Perhaps more explicitly, it looks like `tsctl shownodes up` can only transmit
3983 characters.<br><br>[root@sgate2 ~]# tsctl shownodes up | wc -c<br>3983<br><br>Again, I’m convinced this is a bug not only because the command doesn’t
actually produce a list of all of the up nodes in our cluster; but because
the last name listed is incomplete.<br><br>[root@sgate2 ~]# tsctl shownodes up | tr ',' '\n' | tail -n 1<br>shas0260-opa.rc.int.col[root@sgate2 ~]#<br><br>I’d continue my investigation within tsctl itself but, alas, it’s a binary
with no source code available to me. :)<br><br>I’m trying to get this opened as a bug / PMR; but I’m still working through
the DDN support infrastructure. Thanks for reporting it, though.<br><br>For the record:<br><br>[root@sgate2 ~]# rpm -qa | grep -i gpfs<br>gpfs.base-4.2.1-2.x86_64<br>gpfs.msg.en_US-4.2.1-2.noarch<br>gpfs.gplbin-3.10.0-327.el7.x86_64-4.2.1-0.x86_64<br>gpfs.gskit-8.0.50-57.x86_64<br>gpfs.gpl-4.2.1-2.noarch<br>nfs-ganesha-gpfs-2.3.2-0.ibm24.el7.x86_64<br>gpfs.ext-4.2.1-2.x86_64<br>gpfs.gplbin-3.10.0-327.36.3.el7.x86_64-4.2.1-2.x86_64<br>gpfs.docs-4.2.1-2.noarch<br><br>~jonathon<br><br><br>From: <gpfsug-discuss-bounces@spectrumscale.org> on behalf of Olaf
Weiser <olaf.weiser@de.ibm.com><br>Reply-To: gpfsug main discussion list <gpfsug-discuss@spectrumscale.org><br>Date: Tuesday, January 31, 2017 at 1:30 AM<br>To: gpfsug main discussion list <gpfsug-discuss@spectrumscale.org><br>Subject: Re: [gpfsug-discuss] CES doesn't assign addresses to nodes<br><br>Hi ...same thing here.. everything after 10 nodes will be truncated..<br>though I don't have an issue with it ... I 'll open a PMR .. and I recommend
you to do the same thing.. ;-)<br><br>the reason seems simple.. it is the "| tail" .at the end of the
command.. .. which truncates the output to the last 10 items...<br><br>should be easy to fix..<br>cheers<br>olaf<br><br><br><br><br><br>From: Jonathon A Anderson <jonathon.anderson@colorado.edu><br>To: "gpfsug-discuss@spectrumscale.org"
<gpfsug-discuss@spectrumscale.org><br>Date: 01/30/2017 11:11 PM<br>Subject: Re: [gpfsug-discuss] CES doesn't assign
addresses to nodes<br>Sent by: gpfsug-discuss-bounces@spectrumscale.org<br>________________________________<br><br><br><br><br>In trying to figure this out on my own, I’m relatively certain I’ve found
a bug in GPFS related to the truncation of output from `tsctl shownodes
up`. Any chance someone in development can confirm?<br><br><br>Here are the details of my investigation:<br><br><br>## GPFS is up on sgate2<br><br>[root@sgate2 ~]# mmgetstate<br><br>Node number Node name GPFS state<br>------------------------------------------<br> 414 sgate2-opa active<br><br><br>## but if I tell ces to explicitly put one of our ces addresses on that
node, it says that GPFS is down<br><br>[root@sgate2 ~]# mmces address move --ces-ip 10.225.71.102 --ces-node sgate2-opa<br>mmces address move: GPFS is down on this node.<br>mmces address move: Command failed. Examine previous error messages to
determine cause.<br><br><br>## the “GPFS is down on this node” message is defined as code 109 in
mmglobfuncs<br><br>[root@sgate2 ~]# grep --before-context=1 "GPFS is down on this node."
/usr/lpp/mmfs/bin/mmglobfuncs<br> 109 ) msgTxt=\<br>"%s: GPFS is down on this node."<br><br><br>## and is generated by printErrorMsg in mmcesnetmvaddress when it detects
that the current node is identified as “down” by getDownCesNodeList<br><br>[root@sgate2 ~]# grep --before-context=5 'printErrorMsg 109' /usr/lpp/mmfs/bin/mmcesnetmvaddress<br>downNodeList=$(getDownCesNodeList)<br>for downNode in $downNodeList<br>do<br> if [[ $toNodeName == $downNode ]]<br> then<br> printErrorMsg 109 "$mmcmd"<br><br><br>## getDownCesNodeList is the intersection of all ces nodes with GPFS cluster
nodes listed in `tsctl shownodes up`<br><br>[root@sgate2 ~]# grep --after-context=16 '^function getDownCesNodeList'
/usr/lpp/mmfs/bin/mmcesfuncs<br>function getDownCesNodeList<br>{<br>typeset sourceFile="mmcesfuncs.sh"<br>[[ -n $DEBUG || -n $DEBUGgetDownCesNodeList ]] &&set -x<br>$mmTRACE_ENTER "$*"<br><br>typeset upnodefile=${cmdTmpDir}upnodefile<br>typeset downNodeList<br><br># get all CES nodes<br>$sort -o $nodefile $mmfsCesNodes.dae<br><br>$tsctl shownodes up | $tr ',' '\n' | $sort -o $upnodefile<br><br>downNodeList=$($comm -23 $nodefile $upnodefile)<br>print -- $downNodeList<br>} #----- end of function getDownCesNodeList --------------------<br><br><br>## but not only are the sgate nodes not listed by `tsctl shownodes up`;
its output is obviously and erroneously truncated<br><br>[root@sgate2 ~]# tsctl shownodes up | tr ',' '\n' | tail<br>shas0251-opa.rc.int.colorado.edu<br>shas0252-opa.rc.int.colorado.edu<br>shas0253-opa.rc.int.colorado.edu<br>shas0254-opa.rc.int.colorado.edu<br>shas0255-opa.rc.int.colorado.edu<br>shas0256-opa.rc.int.colorado.edu<br>shas0257-opa.rc.int.colorado.edu<br>shas0258-opa.rc.int.colorado.edu<br>shas0259-opa.rc.int.colorado.edu<br>shas0260-opa.rc.int.col[root@sgate2 ~]#<br><br><br>## I expect that this is a bug in GPFS, likely related to a maximum output
buffer for `tsctl shownodes up`.<br><br><br><br>On 1/24/17, 12:48 PM, "Jonathon A Anderson" <jonathon.anderson@colorado.edu>
wrote:<br><br> I think I'm having the same issue described here:<br><br> </font></tt><a href="http://www.spectrumscale.org/pipermail/gpfsug-discuss/2016-October/002288.html"><tt><font size=3>http://www.spectrumscale.org/pipermail/gpfsug-discuss/2016-October/002288.html</font></tt></a><tt><font size=3><br><br> Any advice or further troubleshooting steps would be much appreciated.
Full disclosure: I also have a DDN case open. (78804)<br><br> We've got a four-node (snsd{1..4}) DDN gridscaler system. I'm trying
to add two CES protocol nodes (sgate{1,2}) to serve NFS.<br><br> Here's the steps I took:<br><br> ---<br> mmcrnodeclass protocol -N sgate1-opa,sgate2-opa<br> mmcrnodeclass nfs -N sgate1-opa,sgate2-opa<br> mmchconfig cesSharedRoot=/gpfs/summit/ces<br> mmchcluster --ccr-enable<br> mmchnode --ces-enable -N protocol<br> mmces service enable NFS<br> mmces service start NFS -N nfs<br> mmces address add --ces-ip 10.225.71.104,10.225.71.105<br> mmces address policy even-coverage<br> mmces address move --rebalance<br> ---<br><br> This worked the very first time I ran it, but the CES addresses
weren't re-distributed after restarting GPFS or a node reboot.<br><br> Things I've tried:<br><br> * disabling ces on the sgate nodes and re-running the above procedure<br> * moving the cluster and filesystem managers to different snsd nodes<br> * deleting and re-creating the cesSharedRoot directory<br><br> Meanwhile, the following log entry appears in mmfs.log.latest every
~30s:<br><br> ---<br> Mon Jan 23 20:31:20 MST 2017: mmcesnetworkmonitor: Found unassigned
address 10.225.71.104<br> Mon Jan 23 20:31:20 MST 2017: mmcesnetworkmonitor: Found unassigned
address 10.225.71.105<br> Mon Jan 23 20:31:20 MST 2017: mmcesnetworkmonitor: handleNetworkProblem
with lock held: assignIP 10.225.71.104_0-_+,10.225.71.105_0-_+ 1<br> Mon Jan 23 20:31:20 MST 2017: mmcesnetworkmonitor: Assigning addresses:
10.225.71.104_0-_+,10.225.71.105_0-_+<br> Mon Jan 23 20:31:20 MST 2017: mmcesnetworkmonitor: moveCesIPs: 10.225.71.104_0-_+,10.225.71.105_0-_+<br> ---<br><br> Also notable, whenever I add or remove addresses now, I see this
in mmsysmonitor.log (among a lot of other entries):<br><br> ---<br> 2017-01-23T20:40:56.363 sgate1 D ET_cesnetwork Entity state without
requireUnique: ces_network_ips_down WARNING No CES relevant NICs detected
- Service.calculateAndUpdateState:275<br> 2017-01-23T20:40:11.364 sgate1 D ET_cesnetwork Update multiple entities
at once {'p2p2': 1, 'bond0': 1, 'p2p1': 1} - Service.setLocalState:333<br> ---<br><br> For the record, here's the interface I expect to get the address
on sgate1:<br><br> ---<br> 11: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 9000
qdisc noqueue state UP<br> link/ether 3c:fd:fe:08:a7:c0 brd ff:ff:ff:ff:ff:ff<br> inet 10.225.71.107/20 brd 10.225.79.255 scope global bond0<br> valid_lft forever preferred_lft forever<br> inet6 fe80::3efd:feff:fe08:a7c0/64 scope link<br> valid_lft forever preferred_lft forever<br> ---<br><br> which is a bond of p2p1 and p2p2.<br><br> ---<br> 6: p2p1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 9000
qdisc mq master bond0 state UP qlen 1000<br> link/ether 3c:fd:fe:08:a7:c0 brd ff:ff:ff:ff:ff:ff<br> 7: p2p2: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 9000
qdisc mq master bond0 state UP qlen 1000<br> link/ether 3c:fd:fe:08:a7:c0 brd ff:ff:ff:ff:ff:ff<br> ---<br><br> A similar bond0 exists on sgate2.<br><br> I crawled around in /usr/lpp/mmfs/lib/mmsysmon/CESNetworkService.py
for a while trying to figure it out, but have been unsuccessful so far.<br><br><br><br>_______________________________________________<br>gpfsug-discuss mailing list<br>gpfsug-discuss at spectrumscale.org<br></font></tt><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss"><tt><font size=3>http://gpfsug.org/mailman/listinfo/gpfsug-discuss</font></tt></a><tt><font size=3><br><br><br><br>_______________________________________________<br>gpfsug-discuss mailing list<br>gpfsug-discuss at spectrumscale.org<br></font></tt><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss"><tt><font size=3>http://gpfsug.org/mailman/listinfo/gpfsug-discuss</font></tt></a><tt><font size=3><br><br><br><br><br><br><br>_______________________________________________<br>gpfsug-discuss mailing list<br>gpfsug-discuss at spectrumscale.org<br></font></tt><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss"><tt><font size=3>http://gpfsug.org/mailman/listinfo/gpfsug-discuss</font></tt></a><tt><font size=3><br></font></tt><br><tt><font size=2>_______________________________________________<br>gpfsug-discuss mailing list<br>gpfsug-discuss at spectrumscale.org<br></font></tt><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss"><tt><font size=2>http://gpfsug.org/mailman/listinfo/gpfsug-discuss</font></tt></a><tt><font size=2><br></font></tt><br><br><BR>