Re: Kernel Tainted

From: Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com>
To: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Kernel Tainted
Date: 2016-10-05 19:43:43
Message-ID: 46b7cf2d-9afb-2943-7996-67d3adbc7e02@2ndquadrant.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 10/05/2016 08:41 PM, reiner peterke wrote:
> Hi,
>
> We are helping a client test an application On Power8 using Postgres
> 9.5.4 which has been compiled specifically for the Power.
>
> This is running on sles12sp1 the current kernel is 3.12.49-11
>
> We are getting these kernel warning associated with the postmaster
> process. The application is handling around 15000TPS It appears that
> one of these messages is generated for each each transaction which fills
> up the warn.log quite quickly.
>
> I’m trying to understand what is causing the Tainted kernel messages.
> the warning is at 'WARNING: at ../net/core/dst.c:287’.
> I’ve found one link that indicates that this is ip6 related.
> https://brunomgalmeida.wordpress.com/2015/07/23/disable-ipv6-postgres-and-pgbouncer/
> Is this accurate? And if these action resolve the error, is it more of
> a bandaid then an actual fix?
>

As Andres already pointed out, this is most likely a kernel issue, not a
PostgreSQL one. The "tainted" has nothing to do with the cause, it's
just a way to inform users whether it's a clean kernel build, or if it
includes code not available in vanilla kernels etc. The "X" means there
are some SuSe-specific modules loaded, IIRC.

And yes, it seems IPv6 related, at least judging by the stack trace:

0xc0000000016f7d80 (unreliable)
sk_dst_check+0x174/0x180
ip6_sk_dst_lookup_flow+0x4c/0x2a0
udpv6_sendmsg+0x688/0xb20
inet_sendmsg+0x9c/0x120
sock_sendmsg+0xec/0x140
SyS_sendto+0x108/0x150
SyS_send+0x50/0x70
SyS_socketcall+0x2a0/0x440
syscall_exit+0x0/0x7c

You should probably talk to SuSe or whoever supports that system.

regards

--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Vitaly Burovoy 2016-10-05 20:05:33 Small doc fix
Previous Message Tom Lane 2016-10-05 19:02:09 Re: Our "fallback" atomics implementation doesn't actually work