Thanks for all your inputs folks. Forcing the switch to operate in 100mbps mode on the said port and then connecting the ROACH1 has solved the problem now.
I am super stumped by the problem - this board worked for close to 10 years auto-negotiating and decided not to, last Friday! Cheers, Ramesh > On 17. Apr 2020, at 16:04, James Smith <jsm...@ska.ac.za> wrote: > > Ramesh, > > I've never actually gotten a ROACH to talk directly to a server. The > auto-negotiation in the NIC is not great. > > You pretty much need a switch in-between. Which is why Jason recommended > using an old 100Mbps switch. > > Regards, > James > > > On Fri, Apr 17, 2020 at 4:00 PM Ramesh Karuppusamy <ram...@mpifr-bonn.mpg.de > <mailto:ram...@mpifr-bonn.mpg.de>> wrote: > > > I just tried this by putting a spare network interface on to 100mbps and also > forcing the ROACH1 eth0 to 100mpbs - this did not help. Looks like the > ethernet device on the ROACH1 is dead. > > I used mii-tool to force 100mbps operation: > > root@roach:~# mii-tool -F 100baseTx-HD eth0 > root@roach:~# mii-tool eth0 > eth0: 100 Mbit, half duplex, link ok > > And similar command on a server with spare eth3 (connected to the ROACH in > question). > > Cheers, > Ramesh > > > > >> On 17. Apr 2020, at 15:41, James Smith <jsm...@ska.ac.za >> <mailto:jsm...@ska.ac.za>> wrote: >> >> Just to add to what Jason has said, older 100 Mbps switches are easy enough >> to pick up secondhand on eBay and such, depending on where you are. They're >> quite cheap. >> >> On Fri, Apr 17, 2020 at 3:19 PM Jason Manley <jman...@ska.ac.za >> <mailto:jman...@ska.ac.za>> wrote: >> >> > Since we are running the ancient Debian Etch on this unit, I have no >> > access to ethtool, with which I understand that you can switch to 100mpbs. >> > Is there any other way to force 100mbps operation? >> >> This actually wouldn't work anyway (we tried it). Something is broken in >> that Nat.Semi PHY's autonegotiation. Sometimes it works, other times not. >> And some boards were more susceptible to failing. >> >> You really need a 100mbps switch, or a 1Gbps switch that you can down-speed >> to lock the port to 100mbps. >> >> Jason >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "casper@lists.berkeley.edu <mailto:casper@lists.berkeley.edu>" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to casper+unsubscr...@lists.berkeley.edu >> <mailto:casper%2bunsubscr...@lists.berkeley.edu>. >> To view this discussion on the web visit >> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/D5D51B6A-2024-4B77-9FC7-7E84CE46DF6F%40ska.ac.za >> >> <https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/D5D51B6A-2024-4B77-9FC7-7E84CE46DF6F%40ska.ac.za>. >> >> -- >> You received this message because you are subscribed to the Google Groups >> "casper@lists.berkeley.edu <mailto:casper@lists.berkeley.edu>" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to casper+unsubscr...@lists.berkeley.edu >> <mailto:casper+unsubscr...@lists.berkeley.edu>. >> To view this discussion on the web visit >> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG67D34keSLJjoTLLjOVHXA0WEcsPbxTyb6p9F10rz%3DPcvcF0w%40mail.gmail.com >> >> <https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG67D34keSLJjoTLLjOVHXA0WEcsPbxTyb6p9F10rz%3DPcvcF0w%40mail.gmail.com?utm_medium=email&utm_source=footer>. > > > -- > You received this message because you are subscribed to the Google Groups > "casper@lists.berkeley.edu <mailto:casper@lists.berkeley.edu>" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to casper+unsubscr...@lists.berkeley.edu > <mailto:casper+unsubscr...@lists.berkeley.edu>. > To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/655FF706-CBDF-4F80-B2EC-BB8B327DB13B%40mpifr-bonn.mpg.de > > <https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/655FF706-CBDF-4F80-B2EC-BB8B327DB13B%40mpifr-bonn.mpg.de?utm_medium=email&utm_source=footer>. > > -- > You received this message because you are subscribed to the Google Groups > "casper@lists.berkeley.edu" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to casper+unsubscr...@lists.berkeley.edu > <mailto:casper+unsubscr...@lists.berkeley.edu>. > To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG67D36OV-KJ%2BY%2BND8sBOCwGWS_HFmUP4fF_cNu7O-0SHhB23g%40mail.gmail.com > > <https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG67D36OV-KJ%2BY%2BND8sBOCwGWS_HFmUP4fF_cNu7O-0SHhB23g%40mail.gmail.com?utm_medium=email&utm_source=footer>. -- You received this message because you are subscribed to the Google Groups "casper@lists.berkeley.edu" group. To unsubscribe from this group and stop receiving emails from it, send an email to casper+unsubscr...@lists.berkeley.edu. To view this discussion on the web visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/A0D9877B-39F1-403A-B83B-4C77F461AE14%40mpifr-bonn.mpg.de.
smime.p7s
Description: S/MIME cryptographic signature