 |
 |
@abh3 requested changes on this pull request.
Looks OK except for two errors and one not so hot variable name. Plus you really should do some tuning on your network :-)
In src/XrdXrootd/XrdXrootdConfig.cc:
> @@ -1441,6 +1443,15 @@ int XrdXrootdProtocol::xmon(XrdOucStream &Config)
if (mrType) monRBval = static_cast<int>(tempval);
else monMBval = static_cast<int>(tempval);
}
+ else if (!strcmp("fbuff", val))
+ {if (!(val = Config.GetWord()))
+ {eDest.Emsg("Config", "monitor fbuff value not specified");
+ return 1;
+ }
+ if (XrdOuca2x::a2sz(eDest,"monitor fbuff", val,
+ &tempval, 1024, 65536)) return 1;
The upper limit should be 65472 not 65536 because it anyone specifies that the packet length in the header will become zero as it's a short. The lower limit should ideally be no less than 4096. If you network drops 4K datagram packets something is really wrong with your network.
In src/XrdXrootd/XrdXrootdMonFile.cc:
> {
// Set the reporting interval and I/O counter
//
repTime = intv;
xfrCnt = xfrcnt;
xfrRem = xfrcnt;
+ fBuff = fbuff;
This isn't going to work at all. If you wind up through the call stack your will notice that Defaults() can be called with fbuff having a value of zero. So this should be:
fBuff = (fbuff <= 0 ? 65472 : fbuff);
Also, why did you call something that is a "size" an fbuff? How about changing it to the more accurate fbsz and fBsz or anything else that tells you its a quantity.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or unsubscribe.
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