Hello all,
I have a 2 node cluster, interconnected via crossed-cable and serial
cable. I noticed that after some time, the serial link went down, with
this message in the kernel log:
kernel: serial8250: too much work for irq4
This is on CentOS 5.2 x86_64 with 2.6.18-92.1.13.el5 and
heartbeat-2.1.4-2.1. Has anyone encountered a similar problem? It's a
Fujitsu Siemens RX300 server. More importantly, is it possible to
reestablish the serial link without takeover/heartbeat restart? (I think
if I kill the ttyS0 hbread and hbwrite processes the other node will
take over).
Thanks very much for answers,
Peter
Here is /proc/interrupts output:
# cat /proc/interrupts
CPU0 CPU1 CPU2 CPU3
0: 1532684436 0 0 0 IO-APIC-edge timer
1: 904 0 2988 0 IO-APIC-edge i8042
4: 7490 0 60869910 0 IO-APIC-edge serial
8: 1 0 0 0 IO-APIC-edge rtc
9: 0 0 0 0 IO-APIC-level acpi
12: 3957 0 24955 0 IO-APIC-edge i8042
98: 152335 0 9145347 0 IO-APIC-level
ehci_hcd:usb1, uhci_hcd:usb2
106: 0 0 0 0 IO-APIC-level
uhci_hcd:usb3
114: 0 0 0 0 IO-APIC-level
uhci_hcd:usb4
122: 22 0 0 0 IO-APIC-level
uhci_hcd:usb5
130: 5261 0 11382292 0 IO-APIC-level ahci
138: 1124609169 0 0 0 PCI-MSI eth0
146: 41138037 0 0 0 PCI-MSI eth1
169: 15451 3060 15725460 291463 IO-APIC-level megasas
NMI: 69963 49649 41985 39739
LOC: 1532683254 1532683175 1532683085 1532683006
ERR: 0
MIS: 0
--
Peter LUCIAK ([email protected])
IBL Software Engineering, http://www.iblsoft.com/
Mierová 103, 82105 Bratislava, Slovakia
Phone: +421-2-32662111, Fax: +421-2-32662110
Direct: +421-2-32662175
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
