> Why not just try the combo updater alone first?
>
> Unless you've already tried it, in which case your only real remaining step 
> is to start over with the DVD and do an archive and install.

Ok, I tried first the 10.5.8 combo, and then the 2010-04 security update.

Finder still went wild.

I unmounted an external HD that had some mounted filesystems on it.

Suddenly, Finder went calm.

I plugged the HD back in. And, checked the system.log:

Jul 12 20:57:58 Kleiman-ibook fseventsd[44]: log dir: /Volumes/WINDOW
DATA/.fseventsd getting new uuid: 2D98D186-BA72-4C9A-B56D-2D9114C1029C
Jul 12 20:57:58 Kleiman-ibook mds[38]: (/)(Error) IndexStore in
get_persistent_id_store:psid: /Volumes/WINDOW
DATA/.Spotlight-V100/Store-V1/Stores/1439A66D-48DA-47C4-A165-6A90D23734F5/psid.db
: danger! ps store table id 497 looks bad! resetting
Jul 12 20:57:59 Kleiman-ibook fseventsd[44]: log dir: /Volumes/DRIVE
SOFT/.fseventsd getting new uuid: B51D25FA-0C63-4166-ADFA-153ED504B108
Jul 12 20:58:00 Kleiman-ibook mds[38]: (/)(Error) IndexStore in
get_persistent_id_store:psid: /Volumes/DRIVE
SOFT/.Spotlight-V100/Store-V1/Stores/77DE9294-99FD-4C49-A610-F991D11563E8/psid.db
: danger! ps store table id 50 looks bad! resetting

This was after a _CLEAN_ shutdown. All 4 partitions (two Fat 32, two
Hfs+) get a quick reindex by spotlight. The two that get complaints in
the log are Fat 32; the two Hfs+ don't get complaints, but do get
reindexed.

Now, the first question is: Since the indexes seem to be corrupted,
what's a safe way to wipe them and force spotlight to do a full
reindex.

The second question is: Since one of those Hfs+ partitions is my Time
Machine backup, and as I understand it, Spotlight and Time Machine go
hand-in-hand, can that index be rebuilt without having to wipe and
restart my time machine?

The third observation/question: One of those Fat32 partitions has
several sparse bundles, each of which is an Hfs+ image. One gets
copies of everything that happens on a Windows XP machine; one
provides a Time Machine backup for that. Neither of them will ever
mount cleanly without an fsck_hfs running. Neither of them every say
why fsck has to run -- nothing shows up in the log. Is this just a
consequence of the drive holding the sparsebundle having a damaged
index, or is this an indication that those also need to be wiped and
start fresh?

Either way, I've definitely determined that the runaway Finder is a
function of this drive being mounted. Now what do I do to fix it?
-- 
Political and economic blog of a strict constitutionalist
http://StrictConstitution.BlogSpot.com
_______________________________________________
MacOSX-talk mailing list
[email protected]
http://www.omnigroup.com/mailman/listinfo/macosx-talk

Reply via email to