I have an application that uses zlib to save all object data to a compressed
stream usinf zlib version 1.1.4.
This stream is being stored by saving it as a file and after an
automation-event, Progress code will read this file and write it to the
database. (transfer via file was much faster than direct transfer)
All works well for more than two years now, except at a few companies where
they operate in a Powerfuse and Citrix environments. It happens once every
month (1 on 1000 sore cycles) that these stored blob data could not be
uncompressed or even the blob has been truncated at random places (and could
also not be decompressed). I could not locate the cause of this problem
My question: Within Powerfuse and Citrix servers it's possible to limit the max
amount of memory and resources for a user. Zlib needs some memory to perform
compression, could this be a problem ?
Does anyone using zlib compression, sometime encounter problems with faulty or
truncated data?
Some data, as long as it is not missing parts of the blob, that returns an
error code (error check generated within the decompressing algoritm) could be
rescued by disabling all decompressing error checks: these results does seen to
be correct.
Any users of zlib (and maybe powerfuse) that does have some ideas on where
these inconsistent data could be generated?
All comments and ideas wil be appreciated.
Best wishes for 2008 and the upcomming holidays.
Regards
Andries Bos
The Netherlands
____________________________________________________________________________________
Never miss a thing. Make Yahoo your home page.
http://www.yahoo.com/r/hs
[Non-text portions of this message have been removed]
