Hi Elvin,
Well, nothing prohibits a kXR_attn from sending a kXR_waitresp. Of course, that would be a silly thing to do and we could just say it’s not allowed in the documentation and if it does happen declare an error. I left the dead code in there because I could not guarantee it wouldn’t be triggered at some point in the future (I guess that was good because of the problem you point out). Anyway, what do you think – prohibit attn+waitresp?
Andy
From: Elvin Sindrilaru
Sent: Tuesday, October 06, 2015 6:27 AM
To: xrootd/xrootd
Cc: Andrew Hanushevsky
Subject: Re: [xrootd] XrdOfsTPCAuth deadlock (#290)
Hi Andy,
After looking over the code, I get you point related to the Ignore flag but this also means that this branch of the switch is dead code: https://github.com/xrootd/xrootd/blob/master/src/XrdCl/XrdClXRootDMsgHandler.cc#L560
Since such a response as you said in the comments is handled synchronously here: https://github.com/xrootd/xrootd/blob/master/src/XrdCl/XrdClStream.cc#L449
All this is true if a kXR_attn response can not contain an embedded kXR_waitreps. Can it? If not, what do you think about removing the dead case branch?
Elvin
—
Reply to this email directly or view it on GitHub.
—
Reply to this email directly or view it on GitHub.
Use REPLY-ALL to reply to list
To unsubscribe from the XROOTD-DEV list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1