Re: [lvs-users] "IPVS: ip_vs_send_async error" flood

To: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Subject: Re: [lvs-users] "IPVS: ip_vs_send_async error" flood
From: Seventh Sven <svensven@xxxxxxxxx>
Date: Wed, 12 May 2010 15:25:54 +0200
On 2010-05-12 13:47, Seventh Sven wrote:
> On 2010-05-12 10:53, Siim Põder wrote:
>> Tomasz Chmielewski wrote:
>>> I have "IPVS: ip_vs_send_async error" literally flooding in dmesg.
> Jumping in with the same problem here. I see the same error on an
> ipvs sync master in a high-volume LVS-DR setup here.
>> Can you check with tcpdump if there are connection sync packets
>> being sent on the wire and how often/how long they are (host
>> and udp port 8848)?
> There are around 100 sync packets sent on every update (every
> second), and they are all 1420 bytes (except the last one, which
> varies). Each bulk is sent within a sub-millisecond window.

Turns out that this is fixed as described by Siim previously in

What I forgot was to restart the sync daemon process. Existing sockets
won't be affected by the increased sysctl value. So:

   net.core.wmem_max = 16777216 # Or tune to your needs
   net.core.rmem_max = 16777216 # Or tune to your needs

..then restart the sync daemon (ipvsadm --stop-daemon master, ipvsadm
--start-daemon backup --mcast-interface <dev>). Same on the backup.

This solved the problem right away.


Please read the documentation before posting - it's available at: mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
or go to

<Prev in Thread] Current Thread [Next in Thread>