Re: Bug in snapshot?

From: Adrian Chadd <adrian@dont-contact.us>
Date: Sun, 23 Apr 2000 06:47:27 +0800

On Sat, Apr 22, 2000, Henrik Nordstrom wrote:
> Reuben Farrelly wrote:
>
> > Ok, I've just gotten your latest patch 21/04 and tried to use it as well -
> > this time it got a lot further but failed again with this message below.
>
> An improvement!
>
> > The fix? Turn off --enable-heap-replacement. I know that Duane knows it
> > is broken in DEVEL2 and obviously it is broken even after your snapshot is
> > applied :-)
>
> > store_dir_ufs.c:543: union has no member named `store_heap'
>
> I haven't tried to make any fixes to the HEAP code, and Adrian is
> changing a lot in the area of replacement policies. But this error is
> probably mostly my fault when I updated the code to match 2.4.DEVEL3
> from 2.3.STABLE1. I'll try to clean it up shortly.

Yup, ditto from here. I'm bughunting a UFS rebuild-from-swaplog bug
where an object is set to be released but it doesnt come out of the
replacement policy (grr), so when that is fixed I'll make sure the
heap replacement code works with UFS.

> > Incidentally, if I turn on --enable-diskd it compiles cleanly. I couldn't
> > get it to work though as it complained about every storage system I entered
> > as being "bungled". What should I be using for this value on my cache_dir
> > line?
>
> The new format for cache_dir is documented in squid.conf:
>
> The change from 2.3 is the introduction of Maxobjsize after the type.
> For ufs you can use -1 for any size. COSS I think requires a size
> limit..
>
> Supported types at this time are probably only ufs and coss (see
> configure --enable-storeio for enabling coss). diskd is currently
> unused, and I don't think asyncufs is fully up to date yet.

diskd will eventually be the async disk IO mechanism squid uses.
--enable-diskd and --disable-diskd shouldn't be applicable in
modio, I'm not sure what is the story in henrik's devel tree.

the maxobjsize is used as the initial hint a to where to place the
object. COSS requires it, and I wouldn't wind it above 100k right now.

Adrian
Received on Sat Apr 22 2000 - 16:47:34 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:12:23 MST