Still getting this issue on this version 9.3.
e1000e 0000:00:19.0 eth0: Detected Hardware Unit Hang:
e1000e 0000:00:19.0 eth0: Reset adapter unexpectedly
This issue has been present since 9.1, when there will be a fix?
e1000e 0000:00:19.0 eth0: Detected Hardware Unit Hang:
e1000e 0000:00:19.0 eth0: Reset adapter unexpectedly
disabling TSO is not enough here. Please disable GRO and GSO for this nic as a workaround:
ethtool -K ethX gso off
ethtool -K ethX gro off
Huh, why announce? I guess this NIC has been addressed together with the other ones, especially those inside the official appliances.
Probably because this NIC was properly identified during the 9.2 Beta (enough info, as requested above).
Agreed, sorry to bump all these different intel nic threads. Just want this to be visible. I have not tried just disabling tso instead of gro and gso, I dont know if its enough or not.
Probably never, if you don't provide the necessary information:
- Exact chip type
- PCI PnP ID (ven & dev), output of "lspci -vn" for NIC
- Output of "ethtool -k ethX"
Owner: Emmanuel Technology Consulting
Former Sophos SG(Astaro) advocate/researcher/Silver Partner
PfSense w/Suricata, ntopng,
Other addons to follow