Hi, I am trying to fix a problem I have for about a week now. The environment is: --several independent servers with Centos5.4 on the latest patchlevel (example will be server2) All of them are working properly. --a machine with Xen installed to host some virtual machines (xenserver1) --domu servers on xenserver1 (crmserver1, winserver1)
The whole network is connected with a Lancom router to a DSL line.
Some days ago I noticed that server2 was updated properly, but xenserver1, crmserver1 and winserver were not updating the OS properly. They seemed to be cut off the internet. I can ping from all to all servers internally, I can ssh from all to all servers internally and I can browse the Internet from server2 successfully (using IP-numbers and hostnames), but xenserver1 and its domu clients could not connect to the internet. I checked the firewalls on xenserver, and crmserver1. They seemed to be OK and stopping them with "etc/init.d/iptables stop" didn't solve the problem.
I can resolve hostnames:
[root@xenserver1 ~]# nslookup www.centos.org Server: 192.168.1.1 Address: 192.168.1.1#53
Non-authoritative answer: Name: www.centos.org Address: 72.232.194.162
but I cannot browse neither to "elinks http://www.centos.org" nor to 72.232.194.162, however, I can ping www.centos.org successfully.
the network routes seem to be OK :
[root@xenserver1 ~]# route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 10.1.1.0 * 255.255.255.0 U 0 0 0 eth1 192.168.122.0 * 255.255.255.0 U 0 0 0 virbr0 169.254.0.0 * 255.255.0.0 U 0 0 0 eth0 192.168.0.0 * 255.255.0.0 U 0 0 0 eth0 default LANCOM_1751_1.p 0.0.0.0 UG 0 0 0 eth0
(eth1 is a drbd interface)
[root@crmserver1 ~]# route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 169.254.0.0 * 255.255.0.0 U 0 0 0 eth0 192.168.0.0 * 255.255.0.0 U 0 0 0 eth0 default LANCOM_1751_1.p 0.0.0.0 UG 0 0 0 eth0
Has anybody any clue where to look?
best regards --- Michael Schumacher PAMAS Partikelmess- und Analysesysteme GmbH Dieselstr.10, D-71277 Rutesheim Tel +49-7152-99630 Fax +49-7152-996333 Geschäftsführer: Gerhard Schreck Handelsregister B Stuttgart HRB 252024
Nobody having an idea?
Hi, I am trying to fix a problem I have for about a week now. The environment is: --several independent servers with Centos5.4 on the latest patchlevel (example will be server2) All of them are working properly. --a machine with Xen installed to host some virtual machines (xenserver1) --domu servers on xenserver1 (crmserver1, winserver1)
The whole network is connected with a Lancom router to a DSL line.
Some days ago I noticed that server2 was updated properly, but xenserver1, crmserver1 and winserver were not updating the OS properly. They seemed to be cut off the internet. I can ping from all to all servers internally, I can ssh from all to all servers internally and I can browse the Internet from server2 successfully (using IP-numbers and hostnames), but xenserver1 and its domu clients could not connect to the internet. I checked the firewalls on xenserver, and crmserver1. They seemed to be OK and stopping them with "etc/init.d/iptables stop" didn't solve the problem.
I can resolve hostnames:
[root@xenserver1 ~]# nslookup www.centos.org Server: 192.168.1.1 Address: 192.168.1.1#53
Non-authoritative answer: Name: www.centos.org Address: 72.232.194.162
but I cannot browse neither to "elinks http://www.centos.org" nor to 72.232.194.162, however, I can ping www.centos.org successfully.
the network routes seem to be OK :
[root@xenserver1 ~]# route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 10.1.1.0 * 255.255.255.0 U 0 0 0 eth1 192.168.122.0 * 255.255.255.0 U 0 0 0 virbr0 169.254.0.0 * 255.255.0.0 U 0 0 0 eth0 192.168.0.0 * 255.255.0.0 U 0 0 0 eth0 default LANCOM_1751_1.p 0.0.0.0 UG 0 0 0 eth0
(eth1 is a drbd interface)
[root@crmserver1 ~]# route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 169.254.0.0 * 255.255.0.0 U 0 0 0 eth0 192.168.0.0 * 255.255.0.0 U 0 0 0 eth0 default LANCOM_1751_1.p 0.0.0.0 UG 0 0 0 eth0
Has anybody any clue where to look?
best regards --- Michael Schumacher
Hello,
Are your netmask config Ok? Can you use wireshark to sniff traffic in the middle LANCOM <->Wireshark/tcpdump<->Xenserver and send me a trace?
Best Regards,
Juan Carlos Díaz Fernández IGN Solutions™ Director Técnico mailto: juan.carlos.diaz@ign-solutions.com callto: +34638295207/+34617413903
Innova Galicia Networks Solutions SL http://www.ign-solutions.com Pazo de Feiras e Congresos de Lugo - Palomar s/n 27004 LUGO - Spain callto: +34982284150 faxto: +34982284922
2010/3/5 Michael Schumacher michael.schumacher@pamas.de
Nobody having an idea?
Hi, I am trying to fix a problem I have for about a week now. The environment is: --several independent servers with Centos5.4 on the latest patchlevel (example will be server2) All of them are working properly. --a machine with Xen installed to host some virtual machines (xenserver1) --domu servers on xenserver1 (crmserver1, winserver1)
The whole network is connected with a Lancom router to a DSL line.
Some days ago I noticed that server2 was updated properly, but xenserver1, crmserver1 and winserver were not updating the OS properly. They seemed to be cut off the internet. I can ping from all to all servers internally, I can ssh from all to all servers internally and I can browse the Internet from server2 successfully (using IP-numbers and hostnames), but xenserver1 and its domu clients could not connect to the internet. I checked the firewalls on xenserver, and crmserver1. They seemed to be OK and stopping them with "etc/init.d/iptables stop" didn't solve the problem.
I can resolve hostnames:
[root@xenserver1 ~]# nslookup www.centos.org Server: 192.168.1.1 Address: 192.168.1.1#53
Non-authoritative answer: Name: www.centos.org Address: 72.232.194.162
but I cannot browse neither to "elinks http://www.centos.org" nor to 72.232.194.162, however, I can ping www.centos.org successfully.
the network routes seem to be OK :
[root@xenserver1 ~]# route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use
Iface
10.1.1.0 * 255.255.255.0 U 0 0 0
eth1
192.168.122.0 * 255.255.255.0 U 0 0 0
virbr0
169.254.0.0 * 255.255.0.0 U 0 0 0
eth0
192.168.0.0 * 255.255.0.0 U 0 0 0
eth0
default LANCOM_1751_1.p 0.0.0.0 UG 0 0 0
eth0
(eth1 is a drbd interface)
[root@crmserver1 ~]# route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use
Iface
169.254.0.0 * 255.255.0.0 U 0 0 0
eth0
192.168.0.0 * 255.255.0.0 U 0 0 0
eth0
default LANCOM_1751_1.p 0.0.0.0 UG 0 0 0
eth0
Has anybody any clue where to look?
best regards
Michael Schumacher
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Dear Juan Carlos,
Are your netmask config Ok? Can you use wireshark to sniff traffic in the middle LANCOM <->Wireshark/tcpdump<->Xenserver and send me a trace?
I couldn't make any further tests on that machine because I couldn't install any software on that machine because of the failing network. However, I took the opportunity and rebooted the machine last night. Guess what, everything is up and running again. Thanks for your assistance and sorry to the list for the disappeared problem. Reboot was no option through the week as this is a production machine under heavy load.
Dear Michael,
I hope your problem won't be a regular issue and you don't see it again.
All of the servers I've installed have a few tools by default to make tests in this cases (tcpdump (wireshark if gui), iptraf, iptables/pf,netcat,sysinternals(windows),...) and have a monitoring packages (Nagios/Ntop) in other machine to obtain data. And one monitor port in our switches to plug my laptop and sniff all traffic on each vlan. I saw a lot of strange issues but if networks and systems have not tools, you only can reboot and try to obtain an answer from logs.
Best Regards,
Juan Carlos Díaz Fernández IGN Solutions™ Director Técnico mailto: juan.carlos.diaz@ign-solutions.com callto: +34638295207/+34617413903
Innova Galicia Networks Solutions SL http://www.ign-solutions.com Pazo de Feiras e Congresos de Lugo - Palomar s/n 27004 LUGO - Spain callto: +34982284150 faxto: +34982284922
2010/3/7 Michael Schumacher michael.schumacher@pamas.de
Dear Juan Carlos,
Are your netmask config Ok? Can you use wireshark to sniff traffic in the middle LANCOM <->Wireshark/tcpdump<->Xenserver and send me a trace?
I couldn't make any further tests on that machine because I couldn't install any software on that machine because of the failing network. However, I took the opportunity and rebooted the machine last night. Guess what, everything is up and running again. Thanks for your assistance and sorry to the list for the disappeared problem. Reboot was no option through the week as this is a production machine under heavy load.
-- Mit freundlichen Grüßen Michael Schumacher mailto:michael.schumacher@pamas.de
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Dear Juan Carlos,
On Monday, March 8, 2010 you wrote:
I hope your problem won't be a regular issue and you don't see it again.
I don't expect to see it again soon. The machine runs rock solid for a year now without any trouble. Years ago, it happened from time to time that services just died after some time (months?), but I think this doesn't happen any longer. I don't know if this comes from the more decent hardware we are using these days or from better software.
All of the servers I've installed have a few tools by default to make tests in this cases (tcpdump (wireshark if gui), iptraf, iptables/pf,netcat,sysinternals(windows),...) and have a monitoring packages (Nagios/Ntop) in other machine to obtain data.
Sounds like a good idea. I will put some tools on the machines.
again, thanks for your help!
best regards --- Michael Schumacher PAMAS Partikelmess- und Analysesysteme GmbH Dieselstr.10, D-71277 Rutesheim Tel +49-7152-99630 Fax +49-7152-996333 Geschäftsführer: Gerhard Schreck Handelsregister B Stuttgart HRB 252024
-
Juan Carlos Díaz Fernández -
Michael Schumacher