Lnf cluster: Difference between revisions
Jump to navigation
Jump to search
No edit summary |
No edit summary |
||
(4 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
Italian cluster (Patrizia Rossi, Federico Ronchetti). Below some information: | Italian cluster (Patrizia Rossi, Federico Ronchetti). Below some information: | ||
Boot sequence: | |||
boot ''lnf1'' first, then ''lnf2'' and ''lnf3''. | |||
Connection scheme: | Connection scheme: | ||
lnf1_eth0 <---> | lnf1_eth0 <---> lnf3_eth1 (cross-over cable) | ||
lnf1_eth1 <---> | lnf1_eth1 <---> lnf2_eth1 (cross_over cable) | ||
lnf1_eth2 <---> switch | lnf1_eth2 <---> switch | ||
lnf2_eth0 <---> switch | |||
lnf3_eth0 <---> switch | |||
Federico Ronchetti wrote: | Federico Ronchetti wrote: |
Latest revision as of 10:09, 24 September 2007
Italian cluster (Patrizia Rossi, Federico Ronchetti). Below some information:
Boot sequence:
boot lnf1 first, then lnf2 and lnf3.
Connection scheme:
lnf1_eth0 <---> lnf3_eth1 (cross-over cable) lnf1_eth1 <---> lnf2_eth1 (cross_over cable) lnf1_eth2 <---> switch lnf2_eth0 <---> switch lnf3_eth0 <---> switch
Federico Ronchetti wrote: > Ciao > > Seems that the mini-farm network is not setup properly. Here is the scheme > > the LNF1 node (NIS server and disk server) has 3 network interfaces. > the first one is public towards the jlab network the remaining two are > private and talk to LNF2 and LNF3. In turn LNF2 and 3 have two network > interfaces each: one public and one private towards LNF1. The > connection of LNF2 and 3 to LNF1 IS DONE WITH CROSSED CABLES. These > seem not to be in place or misconnected. > > In fact you can see from LNF1: > > [root@lnf1 root]# more /etc/hosts > # Do not remove the following line, or various programs > # that require network functionality will fail. > 127.0.0.1 localhost.localdomain localhost > 192.168.1.2 lnf2g.jlab.org lnf2g > 192.168.0.2 lnf3g.jlab.org lnf3g > > and > > [root@lnf1 root]# ping lnf2g > PING lnf2g.jlab.org (192.168.1.2) 56(84) bytes of data. > > --- lnf2g.jlab.org ping statistics --- > 2 packets transmitted, 0 received, 100% packet loss, time 1015ms > > This was done to avoid cluster traffic on the counting house network. > > It's hard to tell where the problem is from here. > > federico > > On 9/21/07, Serguei Boiarinov <boiarino@jlab.org> wrote: > > Hi Patrizia, > > > > as I told you before, you can login as root only to lnf1. > > Good news is that FROM lnf1 you can login as root > > to lnf2 and lnf3, so if something should be fixed it can > > be done remotely. > > > > Access issues have to be resolved between your sysadmin > > and JLAB computer center, I 'm not a person who involve in > > those decisions. > > > > Regards, Sergey > > =================================== > > > > Patrizia Rossi wrote: > > > > > Hi Sergey, > > > > > > still we can not access our mini-farm al jlab. What's the problem? > > > > > > thank you > > > patrizia > > > > > > # Patrizia Rossi > > > # National co-spokesperson of the italian program in Hall B at JLAB (USA) > > > # INFN, Laboratori Nazionali di Frascati - phone/fax +39.06.9403.2549 /2559 > > > # Jefferson Lab, CLAS Collaboration - phone/fax +1.757.269.5284 / 6248 > > > # e-mail:rossi@lnf.infn.it > > > # rossi@jlab.org > > > > > > > > -- > _________________________________________ > > _/ F e d e r i c o R o n c h e t t i > \_ I N F N, Laboratori Nazionali di Frascati > _/ Office Cell Fax:+39.06.9403(2312)(8112)(2559) > \_ google me!! (federico.ronchetti@gmail.com)