Hi,

On our systems I frequently see this kind of log messages for systems
with an Intel 82546GB Ethernet card:

Feb 20 10:12:10 host2 kernel: e1000: eth1: e1000_clean_tx_irq: Detected Tx Unit Hang
Feb 20 10:12:10 host2 kernel:  Tx Queue         <0>
Feb 20 10:12:10 host2 kernel:  TDH            <2c>
Feb 20 10:12:10 host2 kernel:  TDT            <16>
Feb 20 10:12:10 host2 kernel:  next_to_use       <16>
Feb 20 10:12:10 host2 kernel:  next_to_clean     <28>
Feb 20 10:12:10 host2 kernel: buffer_info[next_to_clean]
Feb 20 10:12:10 host2 kernel:  time_stamp       <1060aab38>
Feb 20 10:12:10 host2 kernel:  next_to_watch     <3a>
Feb 20 10:12:10 host2 kernel:  jiffies         <1060ac2e0>
Feb 20 10:12:10 host2 kernel:  next_to_watch.status <0>
Feb 20 10:12:11 host2 kernel: NETDEV WATCHDOG: eth1: transmit timed out
Feb 20 10:12:14 host2 kernel: e1000: eth1: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX

This happens on multiple systems with the same card, so it's not an
individual card defect.

Can anyone comment on what may cause this? Is this a driver problem?

--
--   Jos Vos <jos@(protected)>
--   X/OS Experts in Open Systems BV  |  Phone: +31 20 6938364
--   Amsterdam, The Netherlands     |   Fax: +31 20 6948204

_______________________________________________
rhelv5-list mailing list
rhelv5-list@(protected)
https://www.redhat.com/mailman/listinfo/rhelv5-list