Trevor Smith wrote this: > Why did this happen: Unfortunately; there appears to be a critical timing > within the filesystem/datastore code; if the node is shutdown at just > the right instant the store gets corrupted bad enough that the node > cannot be started again - if I don't wait for my node to get fully up > before I kill it (eg oops I wanted to <something> before starting that) > it seems to have a good chance at toasting itself; but even when the > node is fully up; and just taking it down to adjust the config; turn on > /off stats or otherwise; can corrupt the store as well.
Well, I just got that same error on my Linux-based node today. I've
deleted the store_* file, but have decided not to bother recreating it,
at least for now.
Someone please post a message when the developers get this bug fixed.
('Cause dog knows the developers won't post one.)
--
Greg Wooledge | "Truth belongs to everybody."
[EMAIL PROTECTED] | - The Red Hot Chili Peppers
http://wooledge.org/~greg/ |
msg00773/pgp00000.pgp
Description: PGP signature
