Equiptment:
The problem is identical on the two test machines described below.
1) Sony VIAO notebook 500Mhz, 256MB RAM, Mandrake Linux 8.0, fam 2.6.7
2) Dual 933Mhz Zeon, 512MB ram, Redhat Linux 7.1, fam 2.6.5
Both machines have been upgraded to gcc 3.0.2 but are otherwise the
standard distributions as indicated above.
The Setting:
We have a c program that, given the name of a directory on the command
line, has fam monitor the directory. It then sits on theFAM socket and
waits for messages. all messages received are dumped to the terminal.
When a 'create' or an 'exists' is received for a folder, the program
tells FAM to monitor the directory.
The Problem:
When we decompress (tar xvfz) an archive with a fair number of files
(>50) and folders (>10) into the montored folder, fam does not send a
'create' for all of the new files and folders. It always sends an
'exists' but not always 'created'. Sometimes only the first folder in
the archive gets the 'created' and the rest are 'exists' and other times
1/4 to 1/2 (random?) of the files generate a 'created' the rest only
'exists'. My assumption is that fam cannot keep up, that the whole
thing happens too quickly. Could this be the case? This same
situtation applies if I copy an existing tree (via cp -a) into the
monitored folder.
If you need any additional information, let me know. I can even email
the test program my co-worker wrote if it would be helpful.
--
Source code, list archive, and docs: http://oss.sgi.com/projects/fam/
To unsubscribe: echo unsubscribe fam | mail majordomo@xxxxxxxxxxx
|