Hi again Anirudh, On 12/23/2010 01:16 AM, Michal Ludvig wrote: > Hi Anirudh, > > On 12/21/2010 05:43 AM, anirudh.takkallapa...@emc.com wrote: >> When we start of a download on a big file from S3 using 's3cmd' and >> disconnect the network while it's downloading the file, s3cmd gets >> stuck. Noticed that it gets stuck right in data = >> http_response.read(this_chunk) in S3.py file in the revc_file(...) >> method, guessing this is related to socket time out issue. > To be honest I don't know how exactly the timeout here works. S3cmd uses > the standard Python http library so whatever is the default there will > be the default in s3cmd too. From the quick look in the manual it seems > that only connection timeout can be changed but not a read timeout. I > will check if there is any other way to modify this...
Check out the current s3cmd from SVN trunk - it has a timeout of 10s implemented for socket read/write. You can also change the timeout by setting, e.g. "socket_timeout=33" in your .s3cfg Let me know if it helps. Michal ------------------------------------------------------------------------------ Learn how Oracle Real Application Clusters (RAC) One Node allows customers to consolidate database storage, standardize their database environment, and, should the need arise, upgrade to a full multi-node Oracle RAC database without downtime or disruption http://p.sf.net/sfu/oracle-sfdevnl _______________________________________________ S3tools-general mailing list S3tools-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/s3tools-general
