Reading the docs, it is not clear to me when on-io-error is triggered, i.e. if it depends on disk-timeout or not.
If it does depend on disk-timeout, the warning about post-near-mortem DMA transfers corrupting buffers will probably apply, thus leaving call-local-io-error forcing a reboot the only safe option. I'd like to stop a node before the net timeout elapses, to prevent a dual-primary setup getting a split-brain condition, so waiting for lower levels to signal failure takes too long. Regards, Andreas _______________________________________________ Star us on GITHUB: https://github.com/LINBIT drbd-user mailing list [email protected] http://lists.linbit.com/mailman/listinfo/drbd-user
