Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
> see if this below patch helps. > > -#define GFP_DRBD GFP_ATOMIC > +#define GFP_DRBD GFP_NOIO > Hi Lars - So here's the strange thing. After my last cry for help things seemed to right themselves. All I did was turn off heartbeat altogether. and managed drbd by hand. trying my test from the secondary(made primary) etc. It all seemed to work. Then I rebooted to a clean state on both machines (the only thing changed was that I turned off auto-failback in heartbeat). Everything still worked! Strange. I'm perplexed. I'm assuming that all of this was simply circumstantial and that it would have worked anyway, but the strange thing is that I was able to repeat the problem I had 3 times.. before all of the sudden it started working fine... mysterious. I rebuilt drbd with your patch anyway though and am trying my tests again(so far no scary errors!) :-) So this begs the question of why it would have happened in the first place? Was I having some weird memory usage that I can't replicate now? Have you seen this problem before? Thanks so much for everything btw! Cheers, Tim