Pavel

do you have thi sinstalled? rpm -qa| grep -i TimeDate

If you are on AIX then most probably probably this path missing in etc/stor2rrd.cfg and PERL5LIB, place it at the end: /opt/freeware/lib/perl5/vendor_perl/5.8.8

strange, it normally works, send us logs or try 5 minute Note a short problem description in the text field of the upload form. cd /home/stor2rrd/stor2rrd # or where is your STOR2RRD working dir ps -ef |egrep "2rrd" > tmp/ps_log.txt tar cvhf logs.tar logs etc tmp/*txt gzip -9 logs.tar Send us logs.tar.gz via…

also some useful explanation: http://www.lpar2rrd.com/max_vrs_average.html?4.3.15

hi, it is about averaging, you cannot display all data points from 1 month in the picture with 700 pixles or so. Try zooming in monthly graph, you should see all peaks.

Hi, yes, you can simply change cron schedule. 5 or 10 minuts is generally ok and should work.

Do you mean that data are delayed agout 4 hours for these 2 HMCs? Check HMC time and lslparutil time on this output, are same and actual? su - lpar2rrd cd /home/lpar2rrd/lpar2rrd ./bin/sample_rate.sh

the tool is focused on performance monitoring with some capacity stuff addition. I think that for capacity alerting is better to use native storage tools like Benjamin noted above, each storage has something like that. Would be very hard job to do it for all our supported storages and count with thin provisioned volumes,…

the problem is that there is quite a lot of switches 100+. We read their perf data sequentially from BNA. One run takes 25minutes what is too much. We are working on some paralelization what shoudl resolve issue. We will have a patch soon.

you fix is included in 5.05-2 http://www.lpar2rrd.com/download.htm?4.0

sorry, did not read it via the web, just the first line in email notification I noted. Yes, it is obvious of course.

what means brakes it, data is not imported at all, right?

what graphs exactly you are talking about? Pls screenshot, either here on forum or send it here https://upload.lpar2rrd.com

Hi it is about access to the storage. You say that your storage use only 1750 I do not know if DScli can work only through 1750. Check docu to your DS8k. Open 1750 and try DScli if it is working: /opt/ibm/dscli/dscli -hmc1 192.168.1.1 -user admin here is docu how DScli works, you can prefere 1750 cat…

yes, usually is enough to copy back bin/* only, or config etc/*

vi etc/stor2rrd.cfg and place on start of PERL5LIB variable path (keep rest as it is) PERL5LIB=/usr/lib64/perl5/: ......

let me know when it works, I am also not sure :)

wrong name, sorry find /usr -name RRDp.pm 2>/dev/null

find /usr -name RRDp.pl

Hi, have you installed rrdtool-perl package? rpm -qa| grep -i rrdtool

Hi, thanks! What about to try this code instead? if ( ! $data eq '0xffffffffffffffff' && ishexa($data) == 1 ) { with your change there will be alway presented traffic 4, this should skip it what is more correct. There should be no data == NaNQ Can you test it? It should work as that if it is correctly coded on the daemon…

it will not work, the tool expects unique switch names There is no simple fix from our side, question is if using same switch names is a good practise

Hi, you mean that you have same name of switech accross your fabrics?

There is not visible any switch for user stor2rrd. As per docu "Define All fabrics access in Responsibility Area for user stor2rrd"" Can you see switches when you log in to BNA consiole via stor2rrd user?

It is identified as Solaris based on nmon data. Could you send us nmon file as an example? https://stor2rrd.upload.com

you see it under Solaris? grep lnxhost tmp/menu.txt

It is identified as Linux. On the server side: cd /home/lpar2rrd/lpar2rrd ls -ld data/*/*/lnxhost

Hi, yes, older version do not provide performance data via REST API which we use. It must be V300R003C10 at least how is written in the docu. Sorry for a bit late answer, we have overlooked it.

on the agent side it was ment

Hi, tail -1 /var/tmp/lpar2rrd*txt | cut -d : -f 1-7