Re: Very high interrupt load with rl(4)
Hi, On Thu, 13.08.2009 at 19:24:15 +, Stuart Henderson wrote: > - change the nic; almost anything else would be better I'm seeing rl(4) on these small "embedded" style computers where one can't plug in a regular nic. I don't know how to make vendors ship better interfaces, but can't use very different boxen, either. Kind regards, --Toni++
Re: Very high interrupt load with rl(4)
On 2009-08-13, Wijnand Wiersma wrote: > When investigating I see 80-90% of the CPU time is because of interupts. > Almost all interrupts are for the rl0 and rl1 interface. > > The interrupt load is not coupled to bandwith usage, we can use 60Mbit > without any problems. > I know rl(4) cards are crappy but I am wondering what could cause > these strange spikes? > It must be because of packets per second, but I can't find a offender. almost certainly high PPS. you can watch the traffic with "systat ifs", maybe when you hit a high-load spike you could tcpdump and identify the actual traffic. (or setup pflow and nfdump/etc). > Are there any tweaks to mitigate the effects? > > Right after the load spikes (normally these last for 15 minutes) the > IPSEC tunnels are not in a usable state anymore. > Does anyone have a workaround for this? either of these options would probably help, but might not be (possible|practical): - use a faster cpu - change the nic; almost anything else would be better - convert the rl(4) driver to use MCLGETI - identify the cause of the packet flood and apply lart(8) http://www.xinu.nl/unix/humour/asr-manpages/lart.html
Re: Very high interrupt load with rl(4)
Hi misc, I'me very sorry to reply to the misc list in my native language. The reply was not intended to be send to misc. Did a group reply instead of reply. Sorry. Robert /me will burn in hell On Thu, Aug 13, 2009 at 12:00:40PM +0200, Robert Blacquiere wrote: > Hi Wijnand, > > > On Thu, Aug 13, 2009 at 11:45:43AM +0200, Wijnand Wiersma wrote: > > > > Op 13 aug 2009, om 11:38 heeft Robert Blacquiere het volgende > > geschreven: > > > >> Hi Wijnand, > >> > >> Having fun @ HAR? :) > > Hell yeah! ;-) > > > >> > >> You could try using polling(4) for the rl cards. This could ease > >> interrupts a bit. > >> > > > > There is no polling(4). > > # man polling > > man: no entry for polling in the manual. > > mijn fout :(... Te snel aan het antwoorden... Voor hoge interupts op rl > en andere netwerk kaarten heeft freebsd polling om interupts op dit > soort kaarten te verlangen. Maar je draait openbsd ... mijn fout sorry. > > Veel plezier bij HAR en hopelijk kan iemand anders nog een optie > bedenken voor interupt zaken op openbsd. > > > Groeten > > Robert
Re: Very high interrupt load with rl(4)
Hi Wijnand, On Thu, Aug 13, 2009 at 11:45:43AM +0200, Wijnand Wiersma wrote: > > Op 13 aug 2009, om 11:38 heeft Robert Blacquiere het volgende > geschreven: > >> Hi Wijnand, >> >> Having fun @ HAR? :) > Hell yeah! ;-) > >> >> You could try using polling(4) for the rl cards. This could ease >> interrupts a bit. >> > > There is no polling(4). > # man polling > man: no entry for polling in the manual. mijn fout :(... Te snel aan het antwoorden... Voor hoge interupts op rl en andere netwerk kaarten heeft freebsd polling om interupts op dit soort kaarten te verlangen. Maar je draait openbsd ... mijn fout sorry. Veel plezier bij HAR en hopelijk kan iemand anders nog een optie bedenken voor interupt zaken op openbsd. Groeten Robert
Re: Very high interrupt load with rl(4)
Op 13 aug 2009, om 11:38 heeft Robert Blacquiere het volgende geschreven: Hi Wijnand, Having fun @ HAR? :) Hell yeah! ;-) You could try using polling(4) for the rl cards. This could ease interrupts a bit. There is no polling(4). # man polling man: no entry for polling in the manual. Wijnand
Re: Very high interrupt load with rl(4)
Hi Wijnand, Having fun @ HAR? :) On Thu, Aug 13, 2009 at 11:19:27AM +0200, Wijnand Wiersma wrote: > Hi, > > I am using a machine with only a 500Mhz VIA cpu as NAT/IPSEC gateway > here at the har2009.org camp. > Sometimes the internet is hardly usable, loadavg spikes up to 14. When > investigating I see 80-90% of the CPU time is because of interupts. > Almost all interrupts are for the rl0 and rl1 interface. > > The interrupt load is not coupled to bandwith usage, we can use 60Mbit > without any problems. > I know rl(4) cards are crappy but I am wondering what could cause these > strange spikes? > It must be because of packets per second, but I can't find a offender. > Are there any tweaks to mitigate the effects? > > Right after the load spikes (normally these last for 15 minutes) the > IPSEC tunnels are not in a usable state anymore. > Does anyone have a workaround for this? You could try using polling(4) for the rl cards. This could ease interrupts a bit. Kind regards Robert Blacquiere
Very high interrupt load with rl(4)
Hi, I am using a machine with only a 500Mhz VIA cpu as NAT/IPSEC gateway here at the har2009.org camp. Sometimes the internet is hardly usable, loadavg spikes up to 14. When investigating I see 80-90% of the CPU time is because of interupts. Almost all interrupts are for the rl0 and rl1 interface. The interrupt load is not coupled to bandwith usage, we can use 60Mbit without any problems. I know rl(4) cards are crappy but I am wondering what could cause these strange spikes? It must be because of packets per second, but I can't find a offender. Are there any tweaks to mitigate the effects? Right after the load spikes (normally these last for 15 minutes) the IPSEC tunnels are not in a usable state anymore. Does anyone have a workaround for this? Kind regards, Wijnand