Martin Simmons wrote:
On Mon, 10 Apr 2006 23:12:48 +0200, "Erik P. Olsen" <[EMAIL PROTECTED]> said:
Martin Simmons wrote:
On Sun, 09 Apr 2006 12:16:51 +0200, "Erik P. Olsen" <[EMAIL PROTECTED]> said:
I am infrequently getting tape full conditions at times when I did not expect
it. My tapes are DDS-4 tapes and they are considered "Full" by bacula at various
capacities, for exampel:
21,112,841,569
6,445,884,582
17,477,850,063
15,855,178,172
After the last volume ran "full" and mounting of the next empty volume it
appended only 3,710,244,549 to it, an amount which I had thought could easily be
squeezed into the "full" volume.
I don't understand what triggers the "Full" condition especially the 6GB volume
is wierd since it shouldn't be full before approx. 20GB of data has been written
to it.
I am trying to optimize the use of the tapes but this process is difficult to
carry out when bacula determines the full condition at such different times. Are
there anything I can do to ensure that the full condition becomes close to 20GB
or do I have to live with this strange phenomenon?
Are you using hardware compression?
No, s/w compression. Do you recommend that I use h/w?
H/w compression on DDS-4 is a bit feeble, so it may be better to stick to s/w
compression if you have spare CPU cycles on the fd.
That's the reason I asked, because the h/w compressed size can sometimes be
larger than the uncompressed size, especially if you have both kinds of
compression switched on!
This is very interesting. I have to make my previous statement more precise! I
am NOT using h/w compression at the moment but looking in the calendar I can see
that I actually did use it at the time I had these errors. I dropped h/w
compression because I couldn't see any effect of it. So I am back to the s/w
solution and since then I have not had the error. However, the time is too short
to make a general conclusion. Another problem I've had - also in the h/w
compression period - was system freeze, only cure was power off/power on. It
happened twice and I haven't seen it since.
I'll definitely remain at s/w compression and after a couple of weeks I should
be able to conclude on this issue.
--
Regards,
Erik P. Olsen.
-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
