<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; color: rgb(0, 0, 0); font-size: 14px; font-family: Calibri, sans-serif;">
<div>
<div>
<div>Christopher</div>
<div><br>
</div>
<div>Could you please open a JIRA ticket about this?</div>
<div><br>
</div>
<div>Thanks</div>
<div><br>
</div>
<div>Peter</div>
<div>
<div id="MAC_OUTLOOK_SIGNATURE"></div>
</div>
</div>
</div>
<div><br>
</div>
<span id="OLK_SRC_BODY_SECTION">
<div>
<div>On 8/8/17, 8:58 AM, "lustre-discuss on behalf of Christopher Johnston" <<a href="mailto:lustre-discuss-bounces@lists.lustre.org">lustre-discuss-bounces@lists.lustre.org</a> on behalf of
<a href="mailto:chjohnst@gmail.com">chjohnst@gmail.com</a>> wrote:</div>
</div>
<div><br>
</div>
<blockquote id="MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE" style="BORDER-LEFT: #b5c4df 5 solid; PADDING:0 0 0 5; MARGIN:0 0 0 5;">
<div>
<div>
<div dir="ltr"><span style="color:rgb(0,0,0);font-size:12.8px">At my company we use mmap() exclusively for accessing our data on Lustre. For starters we are seeing some very weird (or maybe expected) poor random read/write performance for these types of access
patterns. I decided to give Lustre 2.10.0 a try with ZFS 0.7.0 as the backend instead of ldisk and after compiling and building the RPMs, the filesystem mounted up just fine. I then started doing some iozone runs to test the stability of the filesystem and
although izone does complete uts benchmark I am seeing a lot of stack traces coming out of various kernel threads. Note I am only seeing this when using mmap(). We also ran our application as well just to verify. I am also going to try with an ldiskfs format
as well to see if this changes anything. </span>
<div style="color:rgb(0,0,0);font-size:12.8px">
<div><br>
</div>
<div>My ZFS settings are modest, with 50% of memory allocated to the ARC:</div>
<div><br>
</div>
<div>options zfs zfs_arc_max=3921674240 zfs_prefetch_disable=1<br>
</div>
<div>recordsize=1M</div>
<div>compression=on</div>
<div>dedupe=off</div>
<div>xattr=sa</div>
<div>dnodesize=auto</div>
<div><br>
</div>
<div><br>
</div>
<div>Below is the output from the stack trace:</div>
<div><br>
</div>
<div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: BUG: Bad page state: 87 messages suppressed</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: BUG: Bad page state in process socknal_sd00_01 pfn:1cbac1</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: page:ffffea00072eb040 count:0 mapcount:-1 mapping: (null) index:0x0</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: page flags: 0x2fffff00008000(tail)</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: page dumped because: nonzero mapcount</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: Modules linked in: 8021q garp mrp stp llc osp(OE) ofd(OE) lfsck(OE) ost(OE) mgc(OE) osd_zfs(OE) lquota(OE) fid(OE) fld(OE) ksocklnd(OE) ptlrpc(OE) obdclass(OE) lnet(OE) libcfs(OE) iosf_mbi crc32_pclmul ghash_clmulni_intel
aesni_intel lrw gf128mul glue_helper ablk_helper cryptd ppdev sg i2c_piix4 parport_pc i2c_core parport pcspkr nfsd nfs_acl lockd grace binfmt_misc auth_rpcgss sunrpc ip_tables xfs libcrc32c zfs(POE) zunicode(POE) zavl(POE) icp(POE) zcommon(POE) znvpair(POE)
spl(OE) zlib_deflate sd_mod crc_t10dif crct10dif_generic virtio_net virtio_scsi crct10dif_pclmul crct10dif_common crc32c_intel serio_raw virtio_pci virtio_ring virtio</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: CPU: 0 PID: 2558 Comm: socknal_sd00_01 Tainted: P B OE ------------ 3.10.0-514.26.2.el7.x86_64 #1</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: ffffea00072eb040 00000000005e265e ffff8800b879f5a8 ffffffff81687133</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: ffff8800b879f5d0 ffffffff81682368 ffffea00072eb040 0000000000000000</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: 000fffff00000000 ffff8800b879f618 ffffffff8118946d fff00000fe000000</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: Call Trace:</div>
<div>Aug 8 09:38:04 dev-gc01-oss001 kernel: [<ffffffff81687133>] dump_stack+0x19/0x1b</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff81682368>] bad_page.part.75+0xdf/0xfc</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8118946d>] free_pages_prepare+0x16d/0x190</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff811897b9>] __free_pages_ok+0x19/0xd0</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8118988b>] free_compound_page+0x1b/0x20</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff81683526>] __put_compound_page+0x1f/0x22</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff81683698>] put_compound_page+0x16f/0x17d</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8118edfc>] put_page+0x4c/0x60</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8155ec1f>] skb_release_data+0x8f/0x140</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8155ecf4>] skb_release_all+0x24/0x30</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8155f1ec>] consume_skb+0x2c/0x80</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8156f06d>] __dev_kfree_skb_any+0x3d/0x50</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffffa0019afb>] free_old_xmit_skbs.isra.32+0x6<wbr>b/0xc0 [virtio_net]</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffffa0019baf>] start_xmit+0x5f/0x4f0 [virtio_net]</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8156f9a1>] dev_hard_start_xmit+0x171/0x3b<wbr>0</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff81597574>] sch_direct_xmit+0x104/0x200</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8157252c>] __dev_queue_xmit+0x23c/0x570</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff81572870>] dev_queue_xmit+0x10/0x20</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815b6876>] ip_finish_output+0x466/0x750</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815b7873>] ip_output+0x73/0xe0</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815b5531>] ip_local_out_sk+0x31/0x40</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815b58a3>] ip_queue_xmit+0x143/0x3a0</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815cf04f>] tcp_transmit_skb+0x4af/0x990</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815cf68a>] tcp_write_xmit+0x15a/0xce0</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815d048e>] __tcp_push_pending_frames+0x2e<wbr>/0xc0</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815bed2c>] tcp_push+0xec/0x120</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815c25b8>] tcp_sendmsg+0xc8/0xc40</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff815ed854>] inet_sendmsg+0x64/0xb0</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff81555ff0>] sock_sendmsg+0xb0/0xf0</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff8168ea3b>] ? _raw_spin_unlock_bh+0x1b/0x40</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff81556067>] kernel_sendmsg+0x37/0x50</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffffa09f40d9>] ksocknal_lib_send_iov+0xd9/0x1<wbr>40 [ksocklnd]</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffffa09ed32f>] ksocknal_process_transmit+0x2a<wbr>f/0xb90 [ksocklnd]</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffffa09f1b84>] ksocknal_scheduler+0x204/0x670 [ksocklnd]</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff810b1b20>] ? wake_up_atomic_t+0x30/0x30</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffffa09f1980>] ? ksocknal_recv+0x2a0/0x2a0 [ksocklnd]</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff810b0a4f>] kthread+0xcf/0xe0</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff810b0980>] ? kthread_create_on_node+0x140/0<wbr>x140</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff81697758>] ret_from_fork+0x58/0x90</div>
<div>Aug 8 09:38:05 dev-gc01-oss001 kernel: [<ffffffff810b0980>] ? kthread_create_on_node+0x140/0<wbr>x140</div>
</div>
</div>
<div style="color:rgb(0,0,0);font-size:12.8px"><br>
</div>
<div style="color:rgb(0,0,0);font-size:12.8px">-Chris</div>
</div>
</div>
</div>
</blockquote>
</span>
</body>
</html>