|
update again, inxi -! 11
04-b1 may now work. I hit an obscure issue with this that made my fixes break, but I finally figured out what is happening. So I believe it will work now. let me know I m ay not get the stable version up soon because I will be pressed for time, but if this works, I'll have stable released at some point in the next few weeks. Back to top |
|||||
|
I'm quite happy thus far though. I really appreciate the effort, Thanks again. Additional Xsigo Systems ((now Oracle) Fabric Director) virtual nics that I added are also visible. See output below:
:: Code ::
[r**t@bigpingu01 ~]# inxi --version inxi 2.1.28-04-b1 (2014-07-22) [root@bigpingu01 ~]# inxi -Nx Network: Card-1: Mellanox MT26428 [ConnectX VPI PCIe 2.0 5GT/s - IB QDR / 10GigE] driver: mlx4_core v: 1.1 bus-ID: 02:00.0 Card-2: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.1 Card-3: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.2 Card-4: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: ec00 Root bus-ID: 05:00.0 Card-5: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.3 Card-6: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: e880 Root bus-ID: 05:00.1 [r**t@bigpingu01 ~]# inxi -ixxxIz Network: Card-1: Mellanox MT26428 [ConnectX VPI PCIe 2.0 5GT/s - IB QDR / 10GigE] driver: mlx4_core v: 1.1 bus-ID: 02:00.0 chip-ID: 15b3:673c IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A Card-2: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.1 chip-ID: 199d:8209 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A Card-3: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.2 chip-ID: 199d:8209 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A Card-4: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: ec00 Root bus-ID: 05:00.0 chip-ID: 8086:10e7 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A Card-5: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.3 chip-ID: 199d:8209 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A Card-6: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: e880 Root bus-ID: 05:00.1 chip-ID: 8086:10e7 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A WAN IP: <filter> IF: bond0 ip: N/A ip-v6: N/A IF: ib0 ip: <filter> ip-v6: <filter> IF: ib1 ip: N/A ip-v6: N/A IF: eth0 ip: N/A ip-v6: <filter> IF: eth1 ip: N/A ip-v6: N/A IF: pvnic0 ip: <filter> ip-v6: <filter> IF: pvnic1 ip: <filter> ip-v6: <filter> IF: vnic0 ip: <filter> ip-v6: <filter> IF: ib0.8100 ip: <filter> ip-v6: <filter> IF: eth0.228@eth0 ip: <filter> ip-v6: <filter> IF: ib0.8110 ip: <filter> ip-v6: N/A IF: ib0.8101 ip: <filter> ip-v6: <filter> Info: Processes: 449 Uptime: 170 days Memory: 1582.2/32099.3MB Init: Upstart v: 0.6.5 runlevel: 3 default: 3 Gcc sys: 4.4.7 Client: Shell (bash 4.1.21 running in tty 1) inxi: 2.1.28-4-b1 [r**t@bigpingu01 ~]# Back to top |
|||||
|
04-b1 had a stupid mistake in it, I fixed it on 05-b1
I I think it will work now. This particular scenario is hard to emulate fully, but I got most of hte issues fixed by just faking the data for dual port single busid devices. inixi -! 11 I think it will work now, let me know. This was tricky because my fix triggered some totally unexpected bash behaviors, which should now be resolveed. That virtual nic I have to see another: inxi -xx@14 with those active to correctly handle them. Back to top |
|||||
|
I removed one of the pvnics. Now with one vnic and one pvnic. I also sent you another @14 with this conf.
:: Code ::
[r**t@bigpingu01 ~]# inxi --version inxi 2.1.28-05-b1 (2014-07-24) [r**t@bigpingu01 ~]# inxi -Nx Network: Card-1: Mellanox MT26428 [ConnectX VPI PCIe 2.0 5GT/s - IB QDR / 10GigE] driver: mlx4_core v: 1.1 bus-ID: 02:00.0 Card-2: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.2 Card-3: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.4 Card-4: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: ec00 Root bus-ID: 05:00.0 Card-5: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: e880 Root bus-ID: 05:00.1 [r**t@bigpingu01 ~]# inxi -ixxxIz Network: Card-1: Mellanox MT26428 [ConnectX VPI PCIe 2.0 5GT/s - IB QDR / 10GigE] driver: mlx4_core v: 1.1 bus-ID: 02:00.0 chip-ID: 15b3:673c IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A Card-2: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.2 chip-ID: 199d:8209 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A Card-3: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.4 chip-ID: 199d:8209 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A Card-4: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: ec00 Root bus-ID: 05:00.0 chip-ID: 8086:10e7 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A Card-5: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: e880 Root bus-ID: 05:00.1 chip-ID: 8086:10e7 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A WAN IP: <filter> IF: bond0 ip: N/A ip-v6: N/A IF: ib0 ip: <filter> ip-v6: <filter> IF: ib1 ip: N/A ip-v6: N/A IF: eth0 ip: N/A ip-v6: <filter> IF: eth1 ip: N/A ip-v6: N/A IF: pvnic0 ip: <filter> ip-v6: <filter> IF: vnic0 ip: <filter> ip-v6: <filter> IF: ib0.8100 ip: <filter> ip-v6: <filter> IF: eth0.228@eth0 ip: <filter> ip-v6: <filter> IF: ib0.8110 ip: <filter> ip-v6: N/A IF: ib0.8101 ip: <filter> ip-v6: <filter> Info: Processes: 449 Uptime: 171 days Memory: 1576.8/32099.3MB Init: Upstart v: 0.6.5 runlevel: 3 default: 3 Gcc sys: 4.4.7 Client: Shell (bash 4.1.21 running in tty 1) inxi: 2.1.28-5-b1 [r**t@bigpingu01 ~]# Back to top |
|||||
|
Ok, sorry about that last one, i didn't test it, the infiniband should now be working with 06-b1, give it a try.
I won't be able to check the virtual device for a day or two, but today's fix should take care of most of the missing data per device id. Let me know if this works, then I'll start on the virtual nic thing. Also, if this fix shows the full data for advanced, -ixxxz then please do another inxi -xx@ 14 as well so I can get the full logging to see what is happening once the main bugs are fixed. Sorry this is taking a while, but these new features are quite tricky to get since I can't do direct testing, but I can emulate certain things as long as I know the full data the systems shows. But today's fix should get most of the original issue fixed. Back to top |
|||||
|
While I believe the infiniband will now work as expected with full -n/-i data, the virtual nics are a different story.
the normal way I get advanced information from those for -n / -i won't work on the virtual nics, I'll take a look at that in a few weeks when I have more time, to see if there's any workarounds possible. We'll see how much data can be reliably and meaningfully assembled about virtual devices, for now, the initial issue you posted on, the infiniband, should be working, please show output for it to confirm. Back to top |
|||||
|
I have sent another @14. Here's the output from inxi 2.1.28-06-b1 (2014-07-25).
:: Code ::
[r**t@bigpingu01 ~]# inxi --version inxi 2.1.28-06-b1 (2014-07-25) [r**t@bigpingu01 ~]# inxi -Nx Network: Card-1: Mellanox MT26428 [ConnectX VPI PCIe 2.0 5GT/s - IB QDR / 10GigE] driver: mlx4_core v: 1.1 bus-ID: 02:00.0 Card-2: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.2 Card-3: Xsigo Systems Virtual NIC Device driver: Xsigo-Virtual-NIC bus-ID: 64:02.4 Card-4: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: ec00 Root bus-ID: 05:00.0 Card-5: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: e880 Root bus-ID: 05:00.1 [r**t@bigpingu01 ~]# inxi -ixxxIz Network: Card-1: Mellanox MT26428 [ConnectX VPI PCIe 2.0 5GT/s - IB QDR / 10GigE] driver: mlx4_core v: 1.1 bus-ID: 02:00.0 chip-ID: 15b3:673c IF: ib0 state: up speed: N/A duplex: N/A mac: <filter> Card-2: Mellanox MT26428 [ConnectX VPI PCIe 2.0 5GT/s - IB QDR / 10GigE] driver: mlx4_core v: 1.1 bus-ID: 02:00.0 chip-ID: 15b3:673c IF: ib1 state: down mac: <filter> Card-3: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: ec00 Root bus-ID: 05:00.0 chip-ID: 8086:10e7 IF: eth0 state: up speed: 1000 Mbps duplex: full mac: <filter> Card-4: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: e880 Root bus-ID: 05:00.1 chip-ID: 8086:10e7 IF: eth1 state: down mac: <filter> Card-5: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: e880 Root bus-ID: 05:00.1 chip-ID: 8086:10e7 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A WAN IP: <filter> IF: bond0 ip: N/A ip-v6: N/A IF: ib0 ip: <filter> ip-v6: <filter> IF: ib1 ip: N/A ip-v6: N/A IF: eth0 ip: N/A ip-v6: <filter> IF: eth1 ip: N/A ip-v6: N/A IF: pvnic0 ip: <filter> ip-v6: <filter> IF: vnic0 ip: <filter> ip-v6: <filter> IF: ib0.8100 ip: <filter> ip-v6: <filter> IF: eth0.228@eth0 ip: <filter> ip-v6: <filter> IF: ib0.8110 ip: <filter> ip-v6: N/A IF: ib0.8101 ip: <filter> ip-v6: <filter> Info: Processes: 449 Uptime: 174 days Memory: 1579.8/32099.3MB Init: Upstart v: 0.6.5 runlevel: 3 default: 3 Gcc sys: 4.4.7 Client: Shell (bash 4.1.21 running in tty 1) inxi: 2.1.28-6-b1 [r**t@bigpingu01 ~]# Back to top |
|||||
|
that's getting better, good. I see a few bugs in inxi in that output, I assume being exposed because of the complexity of that system, which is perfect for bug catching/fixing.
I don't know how much time I will have over the next week, but I'll try to at least locate and fix the bugs I see that are obvious. A few things are pretty hard to fix / make perfect, like the infiniband device with one chip id and one pcibusid showing as two devices, I think I'll leave that because it's too tricky to to figure out a looping logic that would support two ports per pcibusid, but if something comes to me that's clean of course I'll do it. This might be the most perfect system I've ever seen to tighten inxi networking data, thanks for your continued patience in getting all the features working. the main bugs I had in there are now fixed however, I'm just sort of amused to see new ones as well, but the debugging data will show me where that bad data came from in inxi. If you don't hear anything here for a week, it's simply because I don't have time right now, but I'll get the rest fixed within 2 weeks or so, so watch the this thread for updates. There's a few simple things I may fix now however. Back to top |
|||||
|
07-b1 has a small fix, otherwise I think this issue is largely resolved for initial request.
The virtual stuff is another issue which I'll look at next, if I see any way to get that better handled, I'll add that too to next test version, if not, I'll just release what is here as next stable. Please confirm this works as expected, thanks. Fixed was that stray 'Root' in the ports: data, your sample data sets are great for catching subtle little oversight/assumptions glitches like this. The one possible glitch I see from your last -xx@14 inxi output is a doubling of the intel nic 05:00.1: :: Code :: Card-3: Intel 82576 Gigabit Network Connection
driver: igb v: 4.0.1-k ports: ec00 Root bus-ID: 05:00.0 chip-ID: 8086:10e7 IF: eth0 state: up speed: 1000 Mbps duplex: full mac: <edit> Card-4: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: e880 Root bus-ID: 05:00.1 chip-ID: 8086:10e7 IF: eth1 state: down mac: <edit> Card-5: Intel 82576 Gigabit Network Connection driver: igb v: 4.0.1-k ports: e880 Root bus-ID: 05:00.1 chip-ID: 8086:10e7 IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A <update>No, I think I see the cause, the loop isn't fully validating its paths, I see. I'll see if other fixes have handled that, not sure where it's coming from, I don't see anything in the data that could trigger that error. thanks Back to top |
|||||
|
08-b1 I believe will also fix some old bugs in inxi, as well as allowing for the virtual devices etc.
I had to rewrite some very old logic I'd inherited from the first author that was totally bizarre, now I think it works as expected in both cases you show, the multi if single busid device, and the virtual devices. Note that I cannot get any information about the virtual devices beyond what shows now, it's imply not in /sys. Back to top |
|||||
|
All times are GMT - 8 Hours
|
|||||