Thanks for the reply. I upgraded to 1.2.11 but am still getting a seg fault error (it's finding my .rrd file and I've created graphs successfully with similar .rrd file names). I set up the archive to store ~52,000 rows and to expect input every other day. The dump dates (http://tech.teragrid.org/inca/TG/html/rrd_dump) look a little strange though:
<!-- 1990-02-05 04:56:32 PST / 634222592 --> <row><v> NaN </v></row> ... <!-- 2005-08-31 05:56:32 PDT / 1125492992 --> <row><v> NaN </v></row> <!-- 2005-09-02 05:56:32 PDT / 1125665792 --> <row><v> NaN </v></row> ... <!-- 2038-01-18 04:56:32 PST / 2147432192 --> <row><v> NaN </v></row> <!-- 1901-12-14 22:28:16 PST / -2147362304 --> <row><v> NaN </v></row> ... <!-- 2005-08-30 23:28:16 PDT / 1125469696 --> <row><v> NaN </v></row> <!-- 2005-09-01 23:28:16 PDT / 1125642496 --> <row><v> NaN </v></row> <!-- 2005-09-03 23:28:16 PDT / 1125815296 --> <row><v> NaN </v></row> ... <!-- 2038-01-17 22:28:16 PST / 2147408896 --> <row><v> NaN </v></row> <!-- 1901-12-14 16:00:00 PST / -2147385600 --> <row><v> NaN </v></row> ... <!-- 2005-08-30 17:00:00 PDT / 1125446400 --> <row><v> NaN </v></row> <!-- 2005-09-01 17:00:00 PDT / 1125619200 --> <row><v> 0.0000000000e+00 </v></row> <!-- 2005-09-03 17:00:00 PDT / 1125792000 --> <row><v> 0.0000000000e+00 </v></row> The dates wrap around and I have real data at the bottom, but the same dates appear multiple times in the dump. Is it possible this is the problem with graphing? On Mon, Sep 12, 2005 at 06:46:35PM -0700, Kate Ericson wrote: > I'm getting a seg fault error while trying to graph using rrdtool version > 1.0.46 on i386 Linux. I can execute the rrdtool dump command (output at > http://tech.teragrid.org/inca/TG/html/rrd_dump). I think there might be a > problem with my DEF line. Any advice? From: Alex van den Bogaerdt <alex{at}ergens{dot}op{dot}het{dot}net> Date: Tue, 13 Sep 2005 04:44:31 +0200 It shouldn't segfault. That's not right. Is 1.0.46 the latest version? Checking... no, it isn't, 1.0.50 is. Maybe this is a known bug fixed in newer versions. Anyway, unless I'm mistaking, it should produce an error message about a non-existing file. This one to be precise: "tgcp_to_ncsa_home,package=tgcp,service=transfer,serviceType=data,resource=i a64,site=anl,vo=teragrid.rrd" Yes, that's one big filename. Or maybe it selects "tgcp,service=transfer,serviceType=data,resource=ia64,site=anl,vo=teragrid.r rd" or maybe "transfer,serviceType=data,resource=ia64,site=anl,vo=teragrid.rrd" and so on. Also, I'm not sure about the DEFinded datasource name "exitStatus,test" or perhaps "exitStatus,test=tgcp_to_ncsa_home,package" or perhaps "exitStatus,test=tgcp_to_ncsa_home,package=tgcp,service" and so on. Where did you get the idea to use commas and multiple equal signs? What is your intention? > DEF:ncsa_home=exitStatus,test=tgcp_to_ncsa_home,package=tgcp,service=transfe > r,serviceType=data,resource=ia64,site=anl,vo=teragrid.rrd:tgcp:LAST \ If you want datasource "tgcp" from file "teragrid.rrd", the line should read DEF:ncsa_home=teragrid.rrd:tgcp:LAST or DEF:ncsa_home=/path/to/teragrid.rrd:tgcp:LAST Simple: name to use, followed by "=", followed by a filename, followed by a colon, followed by the datasource name you want from that file, etc. If you want multiple datasources, you need to use multiple DEF statements. DEF:var1=/path/to/teragrid.rrd:ds1:LAST DEF:var2=/path/to/teragrid.rrd:ds2:LAST DEF:var3=/path/to/other.rrd:ds1:LAST DEF:var4=/path/to/other.rrd:ds2:LAST and so on. HTH Alex -- Unsubscribe mailto:[EMAIL PROTECTED] Help mailto:[EMAIL PROTECTED] Archive http://lists.ee.ethz.ch/rrd-users WebAdmin http://lists.ee.ethz.ch/lsg2.cgi
