| To: | jamal <hadi@xxxxxxxxxx> |
|---|---|
| Subject: | Re: netlink drops messages. |
| From: | Andi Kleen <ak@xxxxxx> |
| Date: | Thu, 18 Jan 2001 15:24:52 +0100 |
| Cc: | Andi Kleen <ak@xxxxxx>, Werner Almesberger <Werner.Almesberger@xxxxxxx>, "James R. Leu" <jleu@xxxxxxxxxxxxxx>, kuznet@xxxxxxxxxxxxx, gleb@xxxxxxxxxxx, netdev@xxxxxxxxxxx |
| In-reply-to: | <Pine.GSO.4.30.0101180743560.23702-100000@xxxxxxxxxxxxxxxx>; from hadi@xxxxxxxxxx on Thu, Jan 18, 2001 at 01:47:04PM +0100 |
| References: | <20010118125936.B3272@xxxxxxxxxx> <Pine.GSO.4.30.0101180743560.23702-100000@xxxxxxxxxxxxxxxx> |
| Sender: | owner-netdev@xxxxxxxxxxx |
On Thu, Jan 18, 2001 at 01:47:04PM +0100, jamal wrote: > > > On Thu, 18 Jan 2001, Andi Kleen wrote: > > > I suspect if someone is really seriously expecting to handle hundreds of > > interface up/down per seconds they would opt for shared memory. For routes > > you do not even need kernel support, because you can do that privately with > > the routing daemon. > > Sounds nice. I think hundreds of interface up/down per seconds is extreme > end unless you have "dynamic" type of devices like L2TP which come > and go (lets not go into the interface discussion again ;-<). > Having said that, the router has to be robust to hundreds of interface > up/down per seconds. It is when you go to "dump netlink state every 60s" mode on overload. -Andi -- This is like TV. I don't like TV. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: internal drops with tcp, kernel 2.2.16, jamal |
|---|---|
| Next by Date: | Re: internal drops with tcp, kernel 2.2.16, kuznet |
| Previous by Thread: | Re: netlink drops messages., jamal |
| Next by Thread: | Re: netlink drops messages., Benjamin C.R. LaHaise |
| Indexes: | [Date] [Thread] [Top] [All Lists] |