https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6421
Mark Martinec <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Target Milestone|Undefined |3.3.2 --- Comment #1 from Mark Martinec <[email protected]> 2010-04-23 06:09:46 EDT --- So the question is: did spamd fail to send more than 160k of data over SSL, or did spamc fail to receive it past the indicated size. Could you repeat the experiment by using a large message and capture the traffic with 'tcpdump -s 0 -w 0.log ...'. Even though the traffic is encrypted, the size of transfered data should roughly correspond to the size of a nonencrypted data. This could provide the answer to the above question. Glancing over spamd's use of syswrite, I see that it never checks or reports I/O errors, which is really unforgivable. - some calls are to syswrite() directly, other through syswrite_full_buffer; the later should be used throughout as it caters for partial writes and does retries; - even though syswrite_full_buffer() does check for errors, its caller never checks the return value; either the syswrite_full_buffer should be modified to log a warning when all retries fails, or the caller should do so. -- Configure bugmail: https://issues.apache.org/SpamAssassin/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug.
