Hi, On Fri, 2005-04-08 at 04:51, Allen Ziegenfus wrote: > Aha. Upon further inspection I have found the following entries. The > first is repeated many times. Then at the end is the assertion failure > that you rightly postulated. > > Assertion failure in journal_forget() at transaction.c:1217: > "!jh->b_committed_data" Yep, that one is fixed in current kernels. But it's just a symptom of the underlying problem -- the "fix" is to gracefully take the filesystem readonly when we detect this corruption. It doesn't explain why the corruption occurred in the first place. Cheers, Stephen _______________________________________________ Ext3-users@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/ext3-users
