Hi,
I have noticed that when using IE to download a large file via squid stable
11 that the squid process does not revert back to its expected size (roughly
77MB) after completion of the download. The following steps can be used to
recreate the problem:
1. Note the size of the squid process before attempting the file download.
2. Browse to the following web address:
https://jsecom15d.sun.com/ECom/EComActionServlet;jsessionid=F01DECDDF103C000848C92C2DD3434BA
3. Download the file: j2sdk-1_4_2_10-nb-4_1-win-ml.exe. Don’t select save or
run from the IE download dialog, just let squid cache the file.
The output of top will show the squid process size increase as the download
progresses. The cachemgr.cgi page can also be checked to show the increase
in mem_node count. Once the file has been fully cached select save from the
IE download dialog.
I have checked the cachemgr.cgi page and it clearly shows the mem_node count
increase but once the download to the desktop has completed the amount of
allocated memory is still very high.
I have only been able to repeat this “bug” a couple of times and I would not
expect this behaviour as my squid.conf has the memory_pools directive set to
off. Any ideas as to what the problem may be would be a great help.
Thanks in advance,
squid_user
_________________________________________________________________
The new MSN Search Toolbar now includes Desktop search!
http://toolbar.msn.co.uk/
Received on Mon Dec 12 2005 - 08:38:22 MST
This archive was generated by hypermail pre-2.1.9 : Sat Dec 31 2005 - 12:00:02 MST