Subject: Re: [htdig] Replacing db files?
From: David Sklar (email@example.com)
Date: Tue May 23 2000 - 15:50:20 PDT
> Geoff Hutchison:
> At 12:42 PM -0400 5/23/00, David Sklar wrote:
> >have updated indexes. Is there a way I can "lock" htdig so that it will queue
> >up searches or otherwise pause while the files are being copied over? What
> >will the consequences be if htdig runs a search using partially-complete index
> >files that are in process of being copied?
> My suggestion is to transfer the databases under different names
> (e.g. .work files) and then rename them locally. This way you don't
> have to worry about transfer latency and the chance for a bad search
> is minimized.
> If htsearch hits databases in strange states, it will probably give
> some sort of error. There isn't any sort of version system to ensure
> the databases are all of one set. (Hmm, that's not a bad idea...)
I will definitely transfer the database files to the search machines using
different filenames (but on the same partition, so moving them will be quick),
I was just wondering if I should do anything special to htsearch to tell it
not to do searches for the few seconds when the files themselves are actually
being moved or if I need to tell it to not cache anything that it might be
caching because the files have changed, etc..
The context I'm coming from is doing similar things with MySQL where I lock a
table, flush MySQL's buffers, copy new files over existing table files, and
unlock the table.
To unsubscribe from the htdig mailing list, send a message to
You will receive a message to confirm this.
This archive was generated by hypermail 2b28 : Tue May 23 2000 - 13:41:10 PDT