Well, it ultimately _did_ forget about the objects in that bucket. But
only after complaining about each of them many many MANY times (and
about four slow cache-rebuilds...we had box problems that day).
D
Henrik Nordstrom wrote:
>
> Dancer wrote:
>
> > Actually, I have a question of my own (one of my infrequent ones). One
> > of our servers has lost a cache-bucket. Well, the contents of one,
> > anyway. Empty. Nada.
> >
> > It's complaining that it can't open a file in that bucket. About every
> > ten seconds. I presume that it's part of the store-maintenance cycle,
> > but that it would give up and move on.
> >
> > I guess not. So. Anyone got a good idea on how to stop it doing this?
>
> storeSwapInStart should recover after a while.. Each time it encounters
> a swapin failure it should forget about that particular object.
>
> If the directory isn't writeable (no permission, not a directory, out of
> inodes, ...), then you may see storeSwapOutStart errors. I don't think
> Squid recovers automatically in this situation and I beleive this is
> your problem... storeSwapInStart shows the URL, not the swap file name,
> storeSwapOutStart shows the swap file name.
>
> ---
> Henrik Nordstr�m
> Sparetime Squid Hacker
-- Did you read the documentation AND the FAQ? If not, I'll probably still answer your question, but my patience will be limited, and you take the risk of sarcasm and ridicule.Received on Sun Mar 29 1998 - 17:23:17 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:39:29 MST