[RFC PATCH 11/11] afs: Retry rxrpc calls with address rotation on network error

2017-09-01 Thread David Howells
When a network error occurs when we attempt a call, we want to rotate the set of addresses we have for that peer and try the call again. Use the new AF_RXRPC call-retrying facility to do this, thereby avoiding the need to re-encrypt each time as this allows us to reuse the Tx-queue from the dead

[RFC PATCH 11/11] afs: Retry rxrpc calls with address rotation on network error

2017-09-01 Thread David Howells
When a network error occurs when we attempt a call, we want to rotate the set of addresses we have for that peer and try the call again. Use the new AF_RXRPC call-retrying facility to do this, thereby avoiding the need to re-encrypt each time as this allows us to reuse the Tx-queue from the dead

Re: Network error persists in 2.4.4

2001-04-28 Thread Frank de Lange
> (on problems with ne2k-pci on SMP-systems) Seems you're experiencing the effects of the infamous IO-APIC problem ('erratum' in Intel-lingo). There's a patch for these problems by Maciej W. Rozycki, which should (IMnsHO) really be accepted into the main kernel tree since many people are

Network error persists in 2.4.4

2001-04-28 Thread cep
I've had this network bug for the entire to 2.3 dev to 2.4 final and now even in 2.4.4. Whenever I have high network traffic I get an ethernet transmit timeout error by netdev watchdog. This is with a ne2k-pci network card, and smp celeron machine. Reinstalling the module doesn't fix it,

Re: Network error persists in 2.4.4

2001-04-28 Thread Frank de Lange
(on problems with ne2k-pci on SMP-systems) Seems you're experiencing the effects of the infamous IO-APIC problem ('erratum' in Intel-lingo). There's a patch for these problems by Maciej W. Rozycki, which should (IMnsHO) really be accepted into the main kernel tree since many people are

Re: Network error

2000-11-08 Thread Andre Tomt
> Something I seen on a lug. Anyone have a patch for this? > > I'm trying to compile a 2.2.17 kernel. When I do a make bzImage, I get > this error. It seems to be centering on networking areas (nfs, svclock, > tcp, etc.) > > tcp_input.c:1393:52: warning: pasting would not give a valid

Re: Network error

2000-11-08 Thread Carey M. Drake
Guess: you're using RedHat 7.0 (or somehow else are using a "new" version of gcc). Either use make cc=kgcc for redhat or downgrade gcc to a supported version. James Simmons wrote: > > Something I seen on a lug. Anyone have a patch for this? > > I'm trying to compile a 2.2.17 kernel. When I do

Network error

2000-11-08 Thread James Simmons
Something I seen on a lug. Anyone have a patch for this? I'm trying to compile a 2.2.17 kernel. When I do a make bzImage, I get this error. It seems to be centering on networking areas (nfs, svclock, tcp, etc.) tcp_input.c:1393:52: warning: pasting would not give a valid preprocessing token

Network error

2000-11-08 Thread James Simmons
Something I seen on a lug. Anyone have a patch for this? I'm trying to compile a 2.2.17 kernel. When I do a make bzImage, I get this error. It seems to be centering on networking areas (nfs, svclock, tcp, etc.) tcp_input.c:1393:52: warning: pasting would not give a valid preprocessing token

Re: Network error

2000-11-08 Thread Carey M. Drake
Guess: you're using RedHat 7.0 (or somehow else are using a "new" version of gcc). Either use make cc=kgcc for redhat or downgrade gcc to a supported version. James Simmons wrote: Something I seen on a lug. Anyone have a patch for this? I'm trying to compile a 2.2.17 kernel. When I do a

Re: Network error

2000-11-08 Thread Andre Tomt
Something I seen on a lug. Anyone have a patch for this? I'm trying to compile a 2.2.17 kernel. When I do a make bzImage, I get this error. It seems to be centering on networking areas (nfs, svclock, tcp, etc.) tcp_input.c:1393:52: warning: pasting would not give a valid preprocessing