Hi Greg, many thanks for your message here. Am Donnerstag, 23. Mai 2013, 08:04:16 schrieben Sie: > So, the 16384 upper limit may be there because it's not clearly safe to > go higher, and it's rare to need more clusters than that. hmm, depends from usage... ;) > On a system with known memory and workload, certainly you can tweak > settings. I'd try NMBCLUSTERS=32768 and see how that goes. Ok, did that with NetBSD 5.x over several years without problems - will do that again now with 6. > Also, it's possible that you are having a leak. Does this happen more > after long uptimes? No, this only occur's when much network connection load happens (as usual / possible on larger server systems). > which will show the number of denied allocations. Basically, watch > "requests", "fail", and if requests-releases ends up higher and higher. > > One thing to keep in mind is that some network interfaces (e.g. bnx) use > a vast number of clusters, because there are 512 receive slots and at > idle each is filled with an mbuf cluster waiting for an arriving > packet. So a system with 8 bnx interfaces will have 4096 clusters used, > when there is zero traffic. Ok, use xennet interfaces only, but not know if that are affected too. In 5 i experienced a strange prob when using more then one interfaces in different networks - TCP was "scrumbled" / "hanging" on that further networks (reported here before). many thanks for the hints! best regards, Niels. -- --- Niels Dettenbach Syndicat IT & Internet http://www.syndicat.com PGP: https://syndicat.com/pub_key.asc ---
Attachment:
signature.asc
Description: This is a digitally signed message part.