Hi guys, just a quick bit on me: I am still a *nix newbie but am trying to learn the ropes. Since my home server machine only has 1GB of RAM, I use small distributions and have gotten used to doing things on the command line only. My dd-wrt router is set to forward http requests to the VM's proxy ports for transparent proxying. Here's the situation with Squid: My original VM was a downloaded appliance with a stripped down Debian + Squid + Dansguardian + adzapper. I had trouble downloading large files; for example, a file from microsoft.com 1GB in size would stop at around 500MB. This was a problem that was consistently reproduced, though with ~10% variability in filesize, on different Windows Vista/2003 client machines in both FireFox and IE7.0. I created a VM with 128MB RAM and 3GB of hard drive space based on Ubuntu JeOS and installed the following at first: TinyProxy and Dansguardian. I don't think I made any other changes to the dansguardian.conf other than commenting one of the initial lines that would allow it to work. Large files, like the one I originally had a problem with, downloaded fine. (I realize that TinyProxy is different from Squid in that it does not cache.) I was hoping adzapper would work with tinyproxy but there was no such luck. I did "sudo apt-get remove --purge tinyproxy" and then a "sudo apt-get install squid." The Squid installation went without error messages and I believe I made only the following change to squid.conf: "http_port 3128 transparent". The rest of the configuration file was left at default. Adzapper was installed with "sudo apt-get install adzapper". It appears to be working correctly for regular web browsing, but when I try to download large files (this time I tried a 2GB file from microsoft.com), the download incompletes. This brings me back to square one and is very frustrating. I doubt that it is caused by either dansguardian or adzapper. TinyProxy was great, but caching and adzapping is very convenient; I'd hate to see them go. "sudo less /var/log/squid/access.log | grep install.wim" shows only the requests made to GET the large file. That seems to be correct, though I'm not sure if there would be anything remarked by squid when the download fails/stops. Any ideas on how to fix? -- View this message in context: http://www.nabble.com/Trouble-downloading-large-files-with-Squid-tp15277650p15277650.html Sent from the Squid - Users mailing list archive at Nabble.com.
