Bug#925918: linux-image-amd64: linux-image-3.16.0-8-amd64 - unpredictable reboots / kernel panics?
Hi Christian,
On Fri, Apr 19, 2019 at 07:43:14AM +0200, Christian Rusa wrote:
> For me there are still issues with 3.16.0-8-amd64 after updating to
> 3.16.64-2, so I have to still run some machines on 3.16.0-7 to have them
> working.
>
> After boot dmesg gives output like attached.
>
> Software stack which has a problem with this is postfix/rspamd/redis.
> Redis does consume a lot of CPU and postfix is not able to pass mails to
> rspamd using milter interface.
>
> postfix/smtpd[884]: warning: milter inet:localhost:11332: can't read
> SMFIC_OPTNEG reply packet header: Connection timed out
> postfix/smtpd[2999]: warning: milter inet:localhost:11332: read error in
> initial handshake
>
> Downgrading to 3.16.0-7-amd64 3.16.59-1 does fix this issue.
> [ 313.994685] BUG: unable to handle kernel paging request at ffff8823300a1998
> [ 313.994737] IP: [<ffffffff8108b912>] put_pid+0x12/0x50
> [ 313.994776] PGD 1b11067 PUD 0
> [ 313.994799] Oops: 0000 [#1] SMP
> [ 313.994822] Modules linked in: xt_multiport ip6t_REJECT xt_hl ip6t_rt nf_conntrack_ipv6 nf_defrag_ipv6 ipt_REJECT xt_comment xt_LOG xt_limit xt_tcpudp xt_addrtype nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack ip6table_filter ip6_tables nf_conntrack_netbios_ns nf_conntrack_broadcast nf_nat_ftp nf_nat nf_conntrack_ftp nf_conntrack iptable_filter ip_tables x_tables ppdev bochs_drm ttm drm_kms_helper evdev drm joydev pcspkr serio_raw shpchp virtio_balloon parport_pc processor parport thermal_sys button autofs4 ext4 crc16 mbcache jbd2 hid_generic usbhid hid dm_mod virtio_blk virtio_net sg sr_mod cdrom ata_generic psmouse uhci_hcd ehci_hcd virtio_pci virtio_ring virtio i2c_piix4 i2c_core usbcore usb_common floppy ata_piix libata scsi_mod
> [ 313.995268] CPU: 0 PID: 885 Comm: rspamd Not tainted 3.16.0-8-amd64 #1 Debian 3.16.64-2
> [ 313.995310] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.10.2-0-g5f4c7b1-prebuilt.qemu-project.org 04/01/2014
> [ 313.995369] task: ffff8800bbb593d0 ti: ffff88033044c000 task.ti: ffff88033044c000
> [ 313.995407] RIP: 0010:[<ffffffff8108b912>] [<ffffffff8108b912>] put_pid+0x12/0x50
> [ 313.995449] RSP: 0018:ffff88033044fea8 EFLAGS: 00010206
> [ 313.995477] RAX: 0000001ffff10060 RBX: ffff88032f24a140 RCX: 000000000000000e
> [ 313.995514] RDX: 0000000000000000 RSI: ffffffff81848780 RDI: ffff880330191900
> [ 313.995550] RBP: 0000000000002000 R08: ffffffff81848780 R09: 0000000000000000
> [ 313.995586] R10: 0000000000000000 R11: 0000000000000206 R12: ffffffff81889a40
> [ 313.995622] R13: 0000000002000000 R14: 00000000ffffffff R15: ffffffffffffffff
> [ 313.995661] FS: 00007f3c238b37c0(0000) GS:ffff88033fc00000(0000) knlGS:0000000000000000
> [ 313.995702] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 313.995732] CR2: ffff8823300a1998 CR3: 000000032f27c000 CR4: 0000000000000670
> [ 313.995772] Stack:
> [ 313.995786] ffff88032f24a140 ffffffff8123cfb2 ffff88032fc71400 3056535953000019
> [ 313.995830] 0030303030303030 00000000b99a119d 0000000000000000 ffffffff81889b10
> [ 313.995874] ffff88033044ff70 00007f3c17cb17c0 0000000000000000 00007f3c19851680
> [ 313.995918] Call Trace:
> [ 313.995940] [<ffffffff8123cfb2>] ? newseg+0x2b2/0x370
> [ 313.995973] [<ffffffff81237839>] ? ipcget+0xd9/0x1d0
> [ 313.996010] [<ffffffff8123d392>] ? SyS_shmget+0x42/0x50
> [ 313.996044] [<ffffffff8153d15c>] ? system_call_fast_compare_end+0x1c/0x21
> [ 313.996079] Code: 48 c1 e2 05 48 85 c9 48 8b 54 10 38 75 85 0f 1f 00 31 c0 c3 0f 1f 44 00 00 0f 1f 44 00 00 48 85 ff 74 1a 53 8b 47 04 48 c1 e0 05 <48> 8b 5c 07 38 8b 07 83 f8 01 74 12 3e ff 0f 74 0d 5b f3 c3 66
> [ 313.996288] RIP [<ffffffff8108b912>] put_pid+0x12/0x50
> [ 313.996319] RSP <ffff88033044fea8>
> [ 313.996340] CR2: ffff8823300a1998
> [ 313.996361] ---[ end trace ab5bd051ff70cff5 ]---
That looks like a different issue than the root cause of #925918, can
you fill a new bug for it?
In fact the issue might be the same as someone reported on the BeeGFS
user mailinglist at
https://groups.google.com/forum/#!topic/fhgfs-user/DTAVLKSh9g4 .
Regards,
Salvatore
Reply to: