<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">
<div class="">The MDS situation is very basic: active/passive mds0/mds1 for both fas & fsB. fsA has the combined msg/mdt in a single zfs filesystem, and fsB has its own mdt in a separate zfs filesystem. mds0 is primary for all.</div>
<div class=""><br class="">
</div>
<div class="">fsA & fsB DO both have changelogs enabled to feed robinhood databases.</div>
<div class=""><br class="">
</div>
<div class="">What’s the recommended procedure here we should follow before mounting the snapshots?</div>
<div class=""><br class="">
</div>
<div class="">1) disable changelogs on the active mdt’s (this will compromise robinhood, requiring a rescan…), or </div>
<div class="">2) temporarily halt changelog consumption / cleanup (e.g. stop robinhood in our case) and then mount the snapshot?</div>
<div class=""><br class="">
</div>
<div class="">Thanks for the help!
<div class=""><br class="">
</div>
</div>
<div class=""><span style="orphans: 2; widows: 2;" class="">--</span><br class="">
<div class="">
<div class="">
<div style="color: rgb(0, 0, 0); letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">
<div style="orphans: 2; widows: 2;" class="">Benjamin S. Kirk, Ph.D.</div>
<div style="orphans: 2; widows: 2;" class="">NASA Lyndon B. Johnson Space Center</div>
<div style="orphans: 2; widows: 2;" class="">Acting Chief, Aeroscience & Flight Mechanics Division</div>
<div style="orphans: 2; widows: 2;" class=""><br class="">
</div>
</div>
</div>
<div>
<blockquote type="cite" class="">
<div class="">On Aug 27, 2018, at 7:33 PM, Yong, Fan <<a href="mailto:fan.yong@intel.com" class="">fan.yong@intel.com</a>> wrote:</div>
<br class="Apple-interchange-newline">
<div class="">
<div class="">According to the stack trace, someone was trying to cleanup old empty llogs during mount the snapshot. We do NOT allow any modification during mount snapshot; otherwise, it will trigger ZFS backend BUG(). That is why we add LASSERT() when start
the transaction. One possible solution is that, we can add some check in the llog logic to avoid modifying llog under snapshot mode.<br class="">
<br class="">
<br class="">
--<br class="">
Cheers,<br class="">
Nasf<br class="">
<br class="">
-----Original Message-----<br class="">
From: lustre-discuss [<a href="mailto:lustre-discuss-bounces@lists.lustre.org" class="">mailto:lustre-discuss-bounces@lists.lustre.org</a>] On Behalf Of Andreas Dilger<br class="">
Sent: Tuesday, August 28, 2018 5:57 AM<br class="">
To: Kirk, Benjamin (JSC-EG311) <<a href="mailto:benjamin.kirk@nasa.gov" class="">benjamin.kirk@nasa.gov</a>><br class="">
Cc: <a href="mailto:lustre-discuss@lists.lustre.org" class="">lustre-discuss@lists.lustre.org</a><br class="">
Subject: Re: [lustre-discuss] Lustre/ZFS snapshots mount error<br class="">
<br class="">
It's probably best to file an LU ticket for this issue.<br class="">
<br class="">
It looks like there is something with the log processing at mount that is trying to modify the configuration files. I'm not sure whether that should be allowed or not.<br class="">
<br class="">
Does fab have the same MGS as fsA? Does it have the same MDS node as fsA?<br class="">
If it has a different MDS, you might consider to give it its own MGS as well.<br class="">
That doesn't have to be a separate MGS node, just a separate filesystem (ZFS fileset in the same zpool) on the MDS node.<br class="">
<br class="">
Cheers, Andreas<br class="">
<br class="">
<blockquote type="cite" class="">On Aug 27, 2018, at 10:18, Kirk, Benjamin (JSC-EG311) <<a href="mailto:benjamin.kirk@nasa.gov" class="">benjamin.kirk@nasa.gov</a>> wrote:<br class="">
<br class="">
Hi all,<br class="">
<br class="">
We have two filesystems, fsA & fsB (eadc below). Both of which get snapshots taken daily, rotated over a week. It’s a beautiful feature we’ve been using in production ever since it was introduced with 2.10.<br class="">
<br class="">
-) We’ve got Lustre/ZFS 2.10.4 on CentOS 7.5.<br class="">
-) Both fsA & fsB have changelogs active.<br class="">
-) fsA has combined mgt/mdt on a single ZFS filesystem.<br class="">
-) fsB has a single mdt on a single ZFS filesystem.<br class="">
-) for fsA, I have no issues mounting any of the snapshots via lctl.<br class="">
-) for fsB, I can mount the most three recent snapshots, then encounter errors:<br class="">
<br class="">
[root@hpfs-fsl-mds0 ~]# lctl snapshot_mount -F eadc -n eadc_AutoSS-Mon <br class="">
mounted the snapshot eadc_AutoSS-Mon with fsname 3d40bbc<br class="">
[root@hpfs-fsl-mds0 ~]# lctl snapshot_umount -F eadc -n <br class="">
eadc_AutoSS-Mon<br class="">
[root@hpfs-fsl-mds0 ~]# lctl snapshot_mount -F eadc -n eadc_AutoSS-Sun <br class="">
mounted the snapshot eadc_AutoSS-Sun with fsname 584c07a<br class="">
[root@hpfs-fsl-mds0 ~]# lctl snapshot_umount -F eadc -n <br class="">
eadc_AutoSS-Sun<br class="">
[root@hpfs-fsl-mds0 ~]# lctl snapshot_mount -F eadc -n eadc_AutoSS-Sat <br class="">
mounted the snapshot eadc_AutoSS-Sat with fsname 4e646fe<br class="">
[root@hpfs-fsl-mds0 ~]# lctl snapshot_umount -F eadc -n <br class="">
eadc_AutoSS-Sat<br class="">
[root@hpfs-fsl-mds0 ~]# lctl snapshot_mount -F eadc -n eadc_AutoSS-Fri<br class="">
mount.lustre: mount metadata/meta-eadc@eadc_AutoSS-Fri at <br class="">
/mnt/eadc_AutoSS-Fri_MDT0000 failed: Read-only file system Can't mount <br class="">
the snapshot eadc_AutoSS-Fri: Read-only file system<br class="">
<br class="">
The relevant bits from dmesg are<br class="">
[1353434.417762] Lustre: 3d40bbc-MDT0000: set dev_rdonly on this <br class="">
device [1353434.417765] Lustre: Skipped 3 previous similar messages <br class="">
[1353434.649480] Lustre: 3d40bbc-MDT0000: Imperative Recovery enabled, <br class="">
recovery window shrunk from 300-900 down to 150-900 [1353434.649484] <br class="">
Lustre: Skipped 3 previous similar messages [1353434.866228] Lustre: <br class="">
3d40bbc-MDD0000: changelog on [1353434.866233] Lustre: Skipped 1 <br class="">
previous similar message [1353435.427744] Lustre: 3d40bbc-MDT0000: <br class="">
Connection restored to ...@tcp (at ...@tcp) [1353435.427747] Lustre: <br class="">
Skipped 23 previous similar messages [1353445.255899] Lustre: Failing <br class="">
over 3d40bbc-MDT0000 [1353445.255903] Lustre: Skipped 3 previous <br class="">
similar messages [1353445.256150] LustreError: 11-0: <br class="">
3d40bbc-OST0000-osc-MDT0000: operation ost_disconnect to node ...@tcp <br class="">
failed: rc = -107 [1353445.257896] LustreError: Skipped 23 previous <br class="">
similar messages [1353445.353874] Lustre: server umount <br class="">
3d40bbc-MDT0000 complete [1353445.353877] Lustre: Skipped 3 previous <br class="">
similar messages [1353475.302224] Lustre: 4e646fe-MDD0000: changelog <br class="">
on [1353475.302228] Lustre: Skipped 1 previous similar message [1353498.964016] LustreError: 25582:0:(osd_handler.c:341:osd_trans_create()) 36ca26b-MDT0000-osd: someone try to start transaction under readonly mode, should be disabled.<br class="">
[1353498.967260] LustreError: 25582:0:(osd_handler.c:341:osd_trans_create()) Skipped 1 previous similar message<br class="">
[1353498.968829] CPU: 6 PID: 25582 Comm: mount.lustre Kdump: loaded Tainted: P OE ------------ 3.10.0-862.6.3.el7.x86_64 #1<br class="">
[1353498.968830] Hardware name: Supermicro SYS-6027TR-D71FRF/X9DRT, <br class="">
BIOS 3.2a 08/04/2015 [1353498.968832] Call Trace:<br class="">
[1353498.968841] [<ffffffffb5b0e80e>] dump_stack+0x19/0x1b <br class="">
[1353498.968851] [<ffffffffc0cbe5db>] osd_trans_create+0x38b/0x3d0 <br class="">
[osd_zfs] [1353498.968876] [<ffffffffc1116044>] <br class="">
llog_destroy+0x1f4/0x3f0 [obdclass] [1353498.968887] <br class="">
[<ffffffffc111f0f6>] llog_cat_reverse_process_cb+0x246/0x3f0 <br class="">
[obdclass] [1353498.968897] [<ffffffffc111a32c>] <br class="">
llog_reverse_process+0x38c/0xaa0 [obdclass] [1353498.968910] <br class="">
[<ffffffffc111eeb0>] ? llog_cat_process_cb+0x4e0/0x4e0 [obdclass] <br class="">
[1353498.968922] [<ffffffffc111af69>] <br class="">
llog_cat_reverse_process+0x179/0x270 [obdclass] [1353498.968932] <br class="">
[<ffffffffc1115585>] ? llog_init_handle+0xd5/0x9a0 [obdclass] <br class="">
[1353498.968943] [<ffffffffc1116e78>] ? llog_open_create+0x78/0x320 <br class="">
[obdclass] [1353498.968949] [<ffffffffc12e55f0>] ? <br class="">
mdd_root_get+0xf0/0xf0 [mdd] [1353498.968954] [<ffffffffc12ec7af>] <br class="">
mdd_prepare+0x13ff/0x1c70 [mdd] [1353498.968966] [<ffffffffc166b037>] <br class="">
mdt_prepare+0x57/0x3b0 [mdt] [1353498.968983] [<ffffffffc1183afd>] <br class="">
server_start_targets+0x234d/0x2bd0 [obdclass] [1353498.968999] <br class="">
[<ffffffffc1153500>] ? class_config_dump_handler+0x7e0/0x7e0 <br class="">
[obdclass] [1353498.969012] [<ffffffffc118541d>] <br class="">
server_fill_super+0x109d/0x185a [obdclass] [1353498.969025] <br class="">
[<ffffffffc115cef8>] lustre_fill_super+0x328/0x950 [obdclass] <br class="">
[1353498.969038] [<ffffffffc115cbd0>] ? <br class="">
lustre_common_put_super+0x270/0x270 [obdclass] [1353498.969041] <br class="">
[<ffffffffb561f3bf>] mount_nodev+0x4f/0xb0 [1353498.969053] <br class="">
[<ffffffffc1154f18>] lustre_mount+0x38/0x60 [obdclass] <br class="">
[1353498.969055] [<ffffffffb561ff3e>] mount_fs+0x3e/0x1b0 [1353498.969060] [<ffffffffb563d4b7>] vfs_kern_mount+0x67/0x110 [1353498.969062] [<ffffffffb563fadf>] do_mount+0x1ef/0xce0 [1353498.969066] [<ffffffffb55f7c2c>] ? kmem_cache_alloc_trace+0x3c/0x200
[1353498.969069] [<ffffffffb5640913>] SyS_mount+0x83/0xd0 [1353498.969074] [<ffffffffb5b20795>] system_call_fastpath+0x1c/0x21 [1353498.969079] LustreError: 25582:0:(llog_cat.c:1027:llog_cat_reverse_process_cb()) 36ca26b-MDD0000: fail to destroy empty log:
rc = -30<br class="">
[1353498.970785] CPU: 6 PID: 25582 Comm: mount.lustre Kdump: loaded Tainted: P OE ------------ 3.10.0-862.6.3.el7.x86_64 #1<br class="">
[1353498.970786] Hardware name: Supermicro SYS-6027TR-D71FRF/X9DRT, <br class="">
BIOS 3.2a 08/04/2015 [1353498.970787] Call Trace:<br class="">
[1353498.970790] [<ffffffffb5b0e80e>] dump_stack+0x19/0x1b <br class="">
[1353498.970795] [<ffffffffc0cbe5db>] osd_trans_create+0x38b/0x3d0 <br class="">
[osd_zfs] [1353498.970807] [<ffffffffc1117921>] <br class="">
llog_cancel_rec+0xc1/0x880 [obdclass] [1353498.970817] <br class="">
[<ffffffffc111e13b>] llog_cat_cleanup+0xdb/0x380 [obdclass] <br class="">
[1353498.970827] [<ffffffffc111f14d>] <br class="">
llog_cat_reverse_process_cb+0x29d/0x3f0 [obdclass] [1353498.970838] <br class="">
[<ffffffffc111a32c>] llog_reverse_process+0x38c/0xaa0 [obdclass] <br class="">
[1353498.970848] [<ffffffffc111eeb0>] ? <br class="">
llog_cat_process_cb+0x4e0/0x4e0 [obdclass] [1353498.970858] <br class="">
[<ffffffffc111af69>] llog_cat_reverse_process+0x179/0x270 [obdclass] <br class="">
[1353498.970868] [<ffffffffc1115585>] ? llog_init_handle+0xd5/0x9a0 <br class="">
[obdclass] [1353498.970878] [<ffffffffc1116e78>] ? <br class="">
llog_open_create+0x78/0x320 [obdclass] [1353498.970883] <br class="">
[<ffffffffc12e55f0>] ? mdd_root_get+0xf0/0xf0 [mdd] [1353498.970887] <br class="">
[<ffffffffc12ec7af>] mdd_prepare+0x13ff/0x1c70 [mdd] [1353498.970894] <br class="">
[<ffffffffc166b037>] mdt_prepare+0x57/0x3b0 [mdt] [1353498.970908] <br class="">
[<ffffffffc1183afd>] server_start_targets+0x234d/0x2bd0 [obdclass] <br class="">
[1353498.970924] [<ffffffffc1153500>] ? <br class="">
class_config_dump_handler+0x7e0/0x7e0 [obdclass] [1353498.970938] <br class="">
[<ffffffffc118541d>] server_fill_super+0x109d/0x185a [obdclass] <br class="">
[1353498.970950] [<ffffffffc115cef8>] lustre_fill_super+0x328/0x950 <br class="">
[obdclass] [1353498.970962] [<ffffffffc115cbd0>] ? <br class="">
lustre_common_put_super+0x270/0x270 [obdclass] [1353498.970964] <br class="">
[<ffffffffb561f3bf>] mount_nodev+0x4f/0xb0 [1353498.970976] <br class="">
[<ffffffffc1154f18>] lustre_mount+0x38/0x60 [obdclass] <br class="">
[1353498.970978] [<ffffffffb561ff3e>] mount_fs+0x3e/0x1b0 <br class="">
[1353498.970980] [<ffffffffb563d4b7>] vfs_kern_mount+0x67/0x110 <br class="">
[1353498.970982] [<ffffffffb563fadf>] do_mount+0x1ef/0xce0 <br class="">
[1353498.970984] [<ffffffffb55f7c2c>] ? <br class="">
kmem_cache_alloc_trace+0x3c/0x200 [1353498.970986] <br class="">
[<ffffffffb5640913>] SyS_mount+0x83/0xd0 [1353498.970989] <br class="">
[<ffffffffb5b20795>] system_call_fastpath+0x1c/0x21 [1353498.970996] <br class="">
LustreError: 25582:0:(mdd_device.c:354:mdd_changelog_llog_init()) <br class="">
36ca26b-MDD0000: changelog init failed: rc = -30 [1353498.972790] <br class="">
LustreError: 25582:0:(mdd_device.c:427:mdd_changelog_init()) <br class="">
36ca26b-MDD0000: changelog setup during init failed: rc = -30 <br class="">
[1353498.974525] LustreError: <br class="">
25582:0:(mdd_device.c:1061:mdd_prepare()) 36ca26b-MDD0000: failed to <br class="">
initialize changelog: rc = -30 [1353498.976229] LustreError: <br class="">
25582:0:(obd_mount_server.c:1879:server_fill_super()) Unable to start <br class="">
targets: -30 [1353499.072002] LustreError: <br class="">
25582:0:(obd_mount.c:1582:lustre_fill_super()) Unable to mount (-30)<br class="">
<br class="">
<br class="">
I’m hoping those traces mean something to someone - any ideas?<br class="">
<br class="">
Thanks!<br class="">
<br class="">
--<br class="">
Benjamin S. Kirk<br class="">
<br class="">
_______________________________________________<br class="">
lustre-discuss mailing list<br class="">
<a href="mailto:lustre-discuss@lists.lustre.org" class="">lustre-discuss@lists.lustre.org</a><br class="">
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org<br class="">
</blockquote>
<br class="">
Cheers, Andreas<br class="">
---<br class="">
Andreas Dilger<br class="">
CTO Whamcloud<br class="">
<br class="">
<br class="">
<br class="">
<br class="">
</div>
</div>
</blockquote>
</div>
<br class="">
</div>
</div>
</body>
</html>