<div class="socmaildefaultfont" dir="ltr" style="font-family:Arial, Helvetica, sans-serif;font-size:10pt" ><div dir="ltr" >Hi Ryan,</div>
<div dir="ltr" > </div>
<div dir="ltr" >Some interesting IBM-internal communication overnight. The problems seems related to a change made to LINUX_KERNEL_VERSION_VERBOSE to handle the additional digit in the kernel numbering (3.10.0-1000+) . The GPL layer expected LINUX_KERNEL_VERSION_VERBOSE to have that extra digit, and its absence resulted in an incorrect function being compiled in, which led to the crash.</div>
<div dir="ltr" > </div>
<div dir="ltr" >This, at least, seems to make sense, in terms of matching to the symptoms of the problem.</div>
<div dir="ltr" > </div>
<div dir="ltr" >We are still in internal debates on whether/how update our guidelines for gplbin generation ...</div>
<div dir="ltr" > </div>
<div dir="ltr" >Regards,</div>
<div dir="ltr" > </div>
<div dir="ltr" > Felipe</div>
<div dir="ltr" > </div>
<div dir="ltr" >----<br>Felipe Knop knop@us.ibm.com<br>GPFS Development and Security<br>IBM Systems<br>IBM Building 008<br>2455 South Rd, Poughkeepsie, NY 12601<br>(845) 433-9314 T/L 293-9314<br> </div>
<div dir="ltr" > </div>
<div dir="ltr" > </div>
<blockquote data-history-content-modified="1" data-history-expanded="1" dir="ltr" style="border-left:solid #aaaaaa 2px; margin-left:5px; padding-left:5px; direction:ltr; margin-right:0px" >----- Original message -----<br>From: Ryan Novosielski <novosirj@rutgers.edu><br>Sent by: gpfsug-discuss-bounces@spectrumscale.org<br>To: "gpfsug-discuss@spectrumscale.org" <gpfsug-discuss@spectrumscale.org><br>Cc:<br>Subject: [EXTERNAL] Re: [gpfsug-discuss] Kernel BUG/panic in mm/slub.c:3772 on Spectrum Scale Data Access Edition installed via gpfs.gplbin RPM on KVM guests<br>Date: Thu, Jan 16, 2020 4:33 PM<br>
<div><font size="2" face="Default Monospace,Courier New,Courier,monospace" >-----BEGIN PGP SIGNED MESSAGE-----<br>Hash: SHA1<br><br>Hi Felipe,<br><br>I either misunderstood support or convinced them to take further<br>action. It at first looked like they were suggesting "mmbuildgpl fixed<br>it: case closed" (I know they wanted to close the SalesForce case<br>anyway, which would prevent communication on the issue). At this<br>point, they've asked for a bunch more information.<br><br>Support is asking similar questions re: the speculations, and I'll<br>provide them with the relevant output ASAP, but I did confirm all of<br>that, including that there were no stray mmfs26/tracedev kernel<br>modules anywhere else in the relevant /lib/modules PATHs. In the<br>original case, I built on a machine running 3.10.0-957.27.2, but<br>pointed to the 3.10.0-1062.9.1 source code/defined the relevant<br>portions of usr/lpp/mmfs/src/config/env.mcr. That's always worked<br>before, and rebuilding once the build system was running<br>3.10.0-1062.9.1 as well did not change anything either. In all cases,<br>the GPFS version was Spectrum Scale Data Access Edition 5.0.4-1. If<br>you build against either the wrong kernel version or the wrong GPFS<br>version, both will appear right in the filename of the gpfs.gplbin RPM<br>you build. Mine is called:<br><br>gpfs.gplbin-3.10.0-1062.9.1.el7.x86_64-5.0.4-1.x86_64.rpm<br><br>Anyway, thanks for your response; I know you might not be<br>following/working on this directly, but I figured the extra info might<br>be of interest.<br><br>On 1/16/20 8:41 AM, Felipe Knop wrote:<br>> Hi Ryan,<br>><br>> I'm aware of this ticket, and I understand that there has been<br>> active communication with the service team on this problem.<br>><br>> The crash itself, as you indicate, looks like a problem that has<br>> been fixed:<br>><br>> <a href="https://www.ibm.com/support/pages/ibm-spectrum-scale-gpfs-releases-423" target="_blank" >https://www.ibm.com/support/pages/ibm-spectrum-scale-gpfs-releases-423</a><br>13-or-later-and-5022-or-later-have-issues-where-kernel-crashes-rhel76-0<br>><br>> The fact that the problem goes away when *mmbuildgpl* is issued<br>> appears to point to some incompatibility with kernel levels and/or<br>> Scale version levels. Just speculating, some possible areas may<br>> be:<br>><br>><br>> * The RPM might have been built on a version of Scale without the<br>> fix * The RPM might have been built on a different (minor) version<br>> of the kernel * Somehow the VM picked a "leftover" GPFS kernel<br>> module, as opposed to the one included in gpfs.gplbin -- given<br>> that mmfsd never complained about a missing GPL kernel module<br>><br>><br>> Felipe<br>><br>> ---- Felipe Knop knop@us.ibm.com GPFS Development and Security IBM<br>> Systems IBM Building 008 2455 South Rd, Poughkeepsie, NY 12601<br>> (845) 433-9314 T/L 293-9314<br>><br>><br>><br>><br>> ----- Original message ----- From: Ryan Novosielski<br>> <novosirj@rutgers.edu> Sent by:<br>> gpfsug-discuss-bounces@spectrumscale.org To: gpfsug main discussion<br>> list <gpfsug-discuss@spectrumscale.org> Cc: Subject: [EXTERNAL]<br>> [gpfsug-discuss] Kernel BUG/panic in mm/slub.c:3772 on Spectrum<br>> Scale Data Access Edition installed via gpfs.gplbin RPM on KVM<br>> guests Date: Wed, Jan 15, 2020 4:11 PM<br>><br>> Hi there,<br>><br>> I know some of the Spectrum Scale developers look at this list.<br>> I’m having a little trouble with support on this problem.<br>><br>> We are seeing crashes with GPFS 5.0.4-1 Data Access Edition on KVM<br>> guests with a portability layer that has been installed via<br>> gpfs.gplbin RPMs that we built at our site and have used to<br>> install GPFS all over our environment. We’ve not seen this problem<br>> so far on any physical hosts, but have now experienced it on guests<br>> running on number of our KVM hypervisors, across vendors and<br>> firmware versions, etc. At one time I thought it was all happening<br>> on systems using Mellanox virtual functions for Infiniband, but<br>> we’ve now seen it on VMs without VFs. There may be an SELinux<br>> interaction, but some of our hosts have it disabled outright, some<br>> are Permissive, and some were working successfully with 5.0.2.x<br>> GPFS.<br>><br>> What I’ve been instructed to try to solve this problem has been to<br>> run “mmbuildgpl”, and it has solved the problem. I don’t consider<br>> running "mmbuildgpl" a real solution, however. If RPMs are a<br>> supported means of installation, it should work. Support told me<br>> that they’d seen this solve the problem at another site as well.<br>><br>> Does anyone have any more information about this problem/whether<br>> there’s a fix in the pipeline, or something that can be done to<br>> cause this problem that we could remedy? Is there an easy place to<br>> see a list of eFixes to see if this has come up? I know it’s very<br>> similar to a problem that happened I believe it was after 5.0.2.2<br>> and Linux 3.10.0-957.19.1, but that was fixed already in 5.0.3.x.<br>><br>> Below is a sample of the crash output:<br>><br>> [ 156.733477] kernel BUG at mm/slub.c:3772! [ 156.734212] invalid<br>> opcode: 0000 [#1] SMP [ 156.735017] Modules linked in: ebtable_nat<br>> ebtable_filter ebtable_broute bridge stp llc ebtables mmfs26(OE)<br>> mmfslinux(OE) tracedev(OE) rdma_ucm(OE) ib_ucm(OE) rdma_cm(OE)<br>> iw_cm(OE) ib_ipoib(OE) ib_cm(OE) ib_umad(OE) mlx5_fpga_tools(OE)<br>> mlx4_en(OE) mlx4_ib(OE) mlx4_core(OE) ip6table_nat nf_nat_ipv6<br>> ip6table_mangle ip6table_raw nf_conntrack_ipv6 nf_defrag_ipv6<br>> ip6table_filter ip6_tables iptable_nat nf_nat_ipv4 nf_nat<br>> iptable_mangle iptable_raw nf_conntrack_ipv4 nf_defrag_ipv4<br>> xt_comment xt_multiport xt_conntrack nf_conntrack iptable_filter<br>> iptable_security nfit libnvdimm ppdev iosf_mbi crc32_pclmul<br>> ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper<br>> ablk_helper sg joydev pcspkr cryptd parport_pc parport i2c_piix4<br>> virtio_balloon knem(OE) binfmt_misc ip_tables xfs libcrc32c<br>> mlx5_ib(OE) ib_uverbs(OE) ib_core(OE) sr_mod cdrom ata_generic<br>> pata_acpi virtio_console virtio_net virtio_blk crct10dif_pclmul<br>> crct10dif_common mlx5_core(OE) mlxfw(OE) crc32c_intel ptp pps_core<br>> devlink ata_piix serio_raw mlx_compat(OE) libata virtio_pci floppy<br>> virtio_ring virtio dm_mirror dm_region_hash dm_log dm_mod [<br>> 156.754814] CPU: 3 PID: 11826 Comm: request_handle* Tainted: G OE<br>> ------------ 3.10.0-1062.9.1.el7.x86_64 #1 [ 156.756782]<br>> Hardware name: Red Hat KVM, BIOS 1.11.0-2.el7 04/01/2014 [<br>> 156.757978] task: ffff8aeca5bf8000 ti: ffff8ae9f7a24000 task.ti:<br>> ffff8ae9f7a24000 [ 156.759326] RIP: 0010:[<ffffffffbbe23dec>]<br>> [<ffffffffbbe23dec>] kfree+0x13c/0x140 [ 156.760749] RSP:<br>> 0018:ffff8ae9f7a27278 EFLAGS: 00010246 [ 156.761717] RAX:<br>> 001fffff00000400 RBX: ffffffffbc6974bf RCX: ffffa74dc1bcfb60 [<br>> 156.763030] RDX: 001fffff00000000 RSI: ffff8aed90fc6500 RDI:<br>> ffffffffbc6974bf [ 156.764321] RBP: ffff8ae9f7a27290 R08:<br>> 0000000000000014 R09: 0000000000000003 [ 156.765612] R10:<br>> 0000000000000048 R11: ffffdb5a82d125c0 R12: ffffa74dc4fd36c0 [<br>> 156.766938] R13: ffffffffc0a1c562 R14: ffff8ae9f7a272f8 R15:<br>> ffff8ae9f7a27938 [ 156.768229] FS: 00007f8ffff05700(0000)<br>> GS:ffff8aedbfd80000(0000) knlGS:0000000000000000 [ 156.769708] CS:<br>> 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 156.770754] CR2:<br>> 000055963330e2b0 CR3: 0000000325ad2000 CR4: 00000000003606e0 [<br>> 156.772076] DR0: 0000000000000000 DR1: 0000000000000000 DR2:<br>> 0000000000000000 [ 156.773367] DR3: 0000000000000000 DR6:<br>> 00000000fffe0ff0 DR7: 0000000000000400 [ 156.774663] Call Trace: [<br>> 156.775154] [<ffffffffc0a1c562>]<br>> cxiInitInodeSecurityCleanup+0x12/0x20 [mmfslinux] [ 156.776568]<br>> [<ffffffffc0b50562>]<br>> _Z17newInodeInitLinuxP15KernelOperationP13gpfsVfsData_tPP8OpenFilePPvP<br>P10gpfsNode_tP7FileUIDS6_N5LkObj12LockModeEnumE+0x152/0x290<br>><br>><br>[mmfs26]<br>> [ 156.779378] [<ffffffffc0b5cdfa>]<br>> _Z9gpfsMkdirP13gpfsVfsData_tP15KernelOperationP9cxiNode_tPPvPS4_PyS5_P<br>cjjjP10ext_cred_t+0x46a/0x7e0<br>><br>><br>[mmfs26]<br>> [ 156.781689] [<ffffffffc0bdb928>] ?<br>> _ZN14BaseMutexClass15releaseLockHeldEP16KernelSynchState+0x18/0x130<br>><br>><br>[mmfs26]<br>> [ 156.783565] [<ffffffffc0c3db2d>]<br>> _ZL21pcacheHandleCacheMissP13gpfsVfsData_tP15KernelOperationP10gpfsNod<br>e_tPvPcPyP12pCacheResp_tPS5_PS4_PjSA_j+0x4bd/0x760<br>><br>><br>[mmfs26]<br>> [ 156.786228] [<ffffffffc0c40675>]<br>> _Z12pcacheLookupP13gpfsVfsData_tP15KernelOperationP10gpfsNode_tPvPcP7F<br>ilesetjjjPS5_PS4_PyPjS9_+0x1ff5/0x21a0<br>><br>><br>[mmfs26]<br>> [ 156.788681] [<ffffffffc0c023ef>] ?<br>> _Z15findFilesetByIdP15KernelOperationjjPP7Filesetj+0x4f/0xa0<br>> [mmfs26] [ 156.790448] [<ffffffffc0b6d59c>]<br>> _Z10gpfsLookupP13gpfsVfsData_tPvP9cxiNode_tS1_S1_PcjPS1_PS3_PyP10cxiVa<br>ttr_tPjP10ext_cred_tjS5_PiS4_SD_+0x65c/0xad0<br>><br>><br>[mmfs26]<br>> [ 156.793032] [<ffffffffc0b8b022>] ?<br>> _Z33gpfsIsCifsBypassTraversalCheckingv+0xe2/0x130 [mmfs26] [<br>> 156.794588] [<ffffffffc0a36d96>] gpfs_i_lookup+0x2e6/0x5a0<br>> [mmfslinux] [ 156.795838] [<ffffffffc0b6cf40>] ?<br>> _Z8gpfsLinkP13gpfsVfsData_tP9cxiNode_tS2_PvPcjjP10ext_cred_t+0x6c0/0x6<br>c0<br>><br>><br>[mmfs26]<br>> [ 156.797753] [<ffffffffbbe65d52>] ? __d_alloc+0x122/0x180 [<br>> 156.798763] [<ffffffffbbe65e10>] ? d_alloc+0x60/0x70 [<br>> 156.799700] [<ffffffffbbe556d3>] lookup_real+0x23/0x60 [<br>> 156.800651] [<ffffffffbbe560f2>] __lookup_hash+0x42/0x60 [<br>> 156.801675] [<ffffffffbc377874>] lookup_slow+0x42/0xa7 [<br>> 156.802634] [<ffffffffbbe5ac3f>] link_path_walk+0x80f/0x8b0 [<br>> 156.803666] [<ffffffffbbe5ae4a>] path_lookupat+0x7a/0x8b0 [<br>> 156.804690] [<ffffffffbbdcd2fe>] ? lru_cache_add+0xe/0x10 [<br>> 156.805690] [<ffffffffbbe24ef5>] ? kmem_cache_alloc+0x35/0x1f0 [<br>> 156.806766] [<ffffffffbbe5c45f>] ? getname_flags+0x4f/0x1a0 [<br>> 156.807817] [<ffffffffbbe5b6ab>] filename_lookup+0x2b/0xc0 [<br>> 156.808834] [<ffffffffbbe5d5f7>] user_path_at_empty+0x67/0xc0 [<br>> 156.809923] [<ffffffffbbdf3ecd>] ? handle_mm_fault+0x39d/0x9b0 [<br>> 156.811017] [<ffffffffbbe5d661>] user_path_at+0x11/0x20 [<br>> 156.811983] [<ffffffffbbe50343>] vfs_fstatat+0x63/0xc0 [<br>> 156.812951] [<ffffffffbbe506fe>] SYSC_newstat+0x2e/0x60 [<br>> 156.813931] [<ffffffffbc388a26>] ? trace_do_page_fault+0x56/0x150<br>> [ 156.815050] [<ffffffffbbe50bbe>] SyS_newstat+0xe/0x10 [<br>> 156.816010] [<ffffffffbc38dede>] system_call_fastpath+0x25/0x2a [<br>> 156.817104] Code: 49 8b 03 31 f6 f6 c4 40 74 04 41 8b 73 68 4c 89<br>> df e8 89 2f fa ff eb 84 4c 8b 58 30 48 8b 10 80 e6 80 4c 0f 44 d8<br>> e9 28 ff ff ff <0f> 0b 66 90 0f 1f 44 00 00 55 48 89 e5 41 57 41 56<br>> 41 55 41 54 [ 156.822192] RIP [<ffffffffbbe23dec>]<br>> kfree+0x13c/0x140 [ 156.823180] RSP <ffff8ae9f7a27278> [<br>> 156.823872] ---[ end trace 142960be4a4feed8 ]--- [ 156.824806]<br>> Kernel panic - not syncing: Fatal exception [ 156.826475] Kernel<br>> Offset: 0x3ac00000 from 0xffffffff81000000 (relocation range:<br>> 0xffffffff80000000-0xffffffffbfffffff)<br>><br>> -- ____ || \\UTGERS,<br>> |---------------------------*O*--------------------------- ||_//<br>> the State | Ryan Novosielski - novosirj@rutgers.edu || \\<br>> University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS<br>> Campus || \\ of NJ | Office of Advanced Research Computing -<br>> MSB C630, Newark `'<br>><br>> _______________________________________________ gpfsug-discuss<br>> mailing list gpfsug-discuss at spectrumscale.org<br>> <a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss" target="_blank" >http://gpfsug.org/mailman/listinfo/gpfsug-discuss</a><br>><br>><br>><br>><br>><br>> _______________________________________________ gpfsug-discuss<br>> mailing list gpfsug-discuss at spectrumscale.org<br>> <a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss" target="_blank" >http://gpfsug.org/mailman/listinfo/gpfsug-discuss</a><br>><br><br>- --<br> ____<br> || \\UTGERS, |----------------------*O*------------------------<br> ||_// the State | Ryan Novosielski - novosirj@rutgers.edu<br> || \\ University | Sr. Technologist - 973/972.0922 ~*~ RBHS Campus<br> || \\ of NJ | Office of Advanced Res. Comp. - MSB C630, Newark<br> `'<br>-----BEGIN PGP SIGNATURE-----<br><br>iF0EARECAB0WIQST3OUUqPn4dxGCSm6Zv6Bp0RyxvgUCXiDWSgAKCRCZv6Bp0Ryx<br>vpCsAKCQ2ykmeycbOVrHTGaFqb2SsU26NwCg3YyYi4Jy2d+xZjJkE6Vfht8O8gM=<br>=9rKb<br>-----END PGP SIGNATURE-----<br>_______________________________________________<br>gpfsug-discuss mailing list<br>gpfsug-discuss at spectrumscale.org<br><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss" target="_blank" >http://gpfsug.org/mailman/listinfo/gpfsug-discuss</a></font></div></blockquote>
<div dir="ltr" > </div></div><BR>