On Sunday 18 October 2009 01:08:03 Tim Brown wrote:
> On Wednesday 07 October 2009 06:45:09 Felix Wolfsteller wrote:
> > On Tuesday 06 October 2009 23:52:25 Tim Brown wrote:
> Hmm, this is actually a lot more bothersome that I first thought....the
> lifetime of the files seems to be quite long. After generating the filename
> it writes to it and then does daft things like:
>
> files_add_translation (globals, origname, localname);
>
> Using mkstemp alone won't be enough although it did solve the time of check
> time of write which inititially caught my eye :(.
>
> Seems we need to cache the filename we generate, the device ID and the inode
> and check this on each subsequent read but it seems like a real nasty
> solution. Either that or hold the contents in memory and bin the whole write
> to disk element entirely.
>
> Suggestions?
I think, keeping them in memory is the best approach. Thus avoiding any disk IO.
Isn't it an option to have the file contents in the KB? Perhaps define a maximum
length for files and a maximum number of files that can be uploaded?
Best
Jan
--
Dr. Jan-Oliver Wagner | ++49-541-335084-0 | http://www.greenbone.net/
Greenbone Networks GmbH, Neuer Graben 17, 49074 Osnabrück
AG Osnabrück, HR B 202460 | Geschäftsführer: Lukas Grunwald, Dr. Jan-Oliver
Wagner
_______________________________________________
Openvas-devel mailing list
Openvas-devel@wald.intevation.org
http://lists.wald.intevation.org/mailman/listinfo/openvas-devel
