| *** dpanech has quit IRC | 01:29 | |
| *** hyunsikyang__ has quit IRC | 04:45 | |
| *** hyunsikyang has joined #starlingx | 04:51 | |
| *** sgw has quit IRC | 05:55 | |
| *** hyunsikyang__ has joined #starlingx | 07:05 | |
| *** hyunsikyang has quit IRC | 07:09 | |
| *** hyunsikyang has joined #starlingx | 07:13 | |
| *** hyunsikyang__ has quit IRC | 07:17 | |
| *** bengates has joined #starlingx | 07:24 | |
| *** bengates has joined #starlingx | 07:25 | |
| *** yaawang has quit IRC | 07:43 | |
| *** yaawang has joined #starlingx | 07:45 | |
| *** MatrixBridge has joined #starlingx | 09:57 | |
| *** MatrixBridge has left #starlingx | 09:57 | |
| *** MatrixBridge has joined #starlingx | 10:03 | |
| *** MatrixBridge has left #starlingx | 10:03 | |
| *** sysarcher has joined #starlingx | 10:05 | |
| *** MatrixBridge has joined #starlingx | 10:05 | |
| *** MatrixBridge has left #starlingx | 10:05 | |
| *** MatrixBridge has joined #starlingx | 10:06 | |
| *** MatrixBridge has left #starlingx | 10:06 | |
| sysarcher | Hello Folks ... nice to see y'all | 10:13 |
|---|---|---|
| *** riuzen has joined #starlingx | 10:21 | |
| *** riuzen has quit IRC | 10:22 | |
| *** ijolliffe has joined #starlingx | 12:31 | |
| *** stampeder has joined #starlingx | 13:39 | |
| stampeder | Is anyone using the StarlingX iso images for installs? I have found the 13 Dec 2019 3.0.0 fails to install. | 13:40 |
| *** sgw has joined #starlingx | 13:54 | |
| sgw | Morning all | 14:05 |
| stampeder | sgw: Am I in the right place for this question "Is anyone using the StarlingX iso images for installs? I have found the 13 Dec 2019 3.0.0 fails to install." | 14:28 |
| sgw | stampeder: your in the right place, what kind of failure are you seeing? | 14:29 |
| sgw | 3.0 should install on most hardware, what kind of install are you doing? | 14:30 |
| stampeder | Failure of installation script. "Running preinstallation scripts, kickstart script at line 266 fatal, installation aborted". Tried to go to the logs but the installation seems to have changed the root password and centos nor sysadmin works. | 14:31 |
| stampeder | Just a further piece of info. I previously downloaded the daily image from 04-16-20 and it worked flawlessly. It has other issues and that is why I was trying to install the 13 dec image and hit the install failure. | 14:39 |
| stampeder | I'm a network engineer and not a developer. I write code to get my job done and leave the difficult stuff to others. | 14:40 |
| *** sysarcher has quit IRC | 14:52 | |
| *** rcw has joined #starlingx | 14:54 | |
| dpenney__ | ctrl-o in the console window where you see the installation failure will get you to a shell, where you can see Anaconda logs in /tmp. Any more information provided by the error? | 15:05 |
| sgw | stampeder: sorry was on a call | 15:06 |
| stampeder | dpenny: will fire it up again and do the ctrl-o. Will let you know the info in /tmp | 15:08 |
| stampeder | OK. Got some more informaiton on the error. Running.../bin/sh /tmp/ks-script-jCY194 Installation failed ERROR: Specified installation (sda) or boot (sda) device is a usb drive. Debug program: Return code: 1. So for some reason it is not seeing my hard drive for install. It is stuck on the USB stick for install. | 15:35 |
| stampeder | I have a 1T hard disk on the system. Running df fails to see the 1T drive. | 15:38 |
| dpenney__ | but it worked fine with the 04-16-20 load? I wonder if it's a newer disk that is not recognized by the kernel in the 3.0 ISO | 15:42 |
| sgw | stampeder: what kind of drive do you have maybe it's an NVME? Oh but you said master worked | 15:43 |
| stampeder | No. Master did not work. Gives the same install errors. Daily build from 04-16 installed perfectly. I was able to get controller-0 working and unlocked. So it's not the drive. In fact because these images don't install I can reboot to the 04-16 image on the 1T drive after I remove the USB. | 15:46 |
| stampeder | Weird. | 15:46 |
| sgw | So it's a 1T SSD or NVME? | 15:47 |
| stampeder | Quick aside. When starlingx is up and running where will I find the release number. I didn't see an equivalent to /etc/os-release, and I don't think uname will work. The drive is not a SSD. It is NVME. Spinning disk. Western Digital Green WD10EADS. | 15:49 |
| sgw | HDD then nvme is another form of memory device | 15:49 |
| sgw | There is an /etc/build-info file that contains additional information about the build | 15:49 |
| sgw | What /dev/sd* entry is the Green drive? | 15:50 |
| dpenney__ | load version is in /etc/build.info... build ID and date | 15:50 |
| sgw | sorry . not -! | 15:51 |
| dpenney__ | my guess would be that the drive is not supported by the older kernel that's in the 3.0 ISO | 15:52 |
| stampeder | The drive is showing up as sdc in /dev. | 15:52 |
| sgw | Yup that's why! We use sda or sdb as target install, do you have another drive or USB stick in the machine? | 15:53 |
| sgw | dpenney__: that hdd should just be a standard sd* device, nothing special about it. | 15:53 |
| dpenney__ | sgw: ok | 15:54 |
| dpenney__ | so the USB is getting detected first and assigned the sda, which is default... you can edit the boot cmdline to set "rootfs_device=sdc boot_device=sdc" ... odd that the ordering would be different in the newer load | 15:55 |
| sgw | I wonder what's being detected as sdb though | 15:56 |
| stampeder | Oh. Interesting thought. I had another drive in the box temporarily while I used Clonezilla to back it up. I have since removed the backup drive. So the HDD would have shown up as sda and the clonezilla backup would have shown up as sdb. Now the hdd is stuck at being sda but the usb is also sda. Hmmm... | 15:56 |
| sgw | Check your BIOS also | 15:57 |
| stampeder | No. Check that. The hdd is showing up as sdc. | 15:57 |
| stampeder | I'll go check the BIOS. | 15:57 |
| sgw | once you remove the other drive you should only have 2 drives sda / sdb | 15:57 |
| sgw | BIOS does not remember empty slots | 15:57 |
| stampeder | I just checked the bios and it only sees the one hdd drive. | 16:00 |
| stampeder | My bios has a feature that forces a check for usb stick on boot and then boots it automatically. However, it shouldn't result in the hdd showing up as sdc. | 16:01 |
| sgw | Yeah, not sure about that. sorry, but as dpenney__ suggests you can modify the boot cmdline in grub | 16:03 |
| stampeder | In the first part of the install just before we get to random: crng init done the hdd shows up as sdb. | 16:04 |
| sgw | jsomervi1: I think I found a solution to the tarball download vs rpm download, you can see my latest patch I use the .src.rpm as the tarball name that way if the .src.rpm is not available it will get rebuilt and use the into_srpm.sh to download the correct tarball. The PRC folks tested it both Intel and Neusoft, I think it's good to go. | 16:06 |
| *** bengates has quit IRC | 16:10 | |
| stampeder | I"ve checked the bios and rebooted. It failed at the same spot line 266. looking at /dev I see sda and sdb. The error says the install failed because sda is a USB device. Is it possible that the bios by forcing the boot when a usb is present is causing the confusion and not allowing the script to see sdb as the install disk? I know it sounds weird | 16:17 |
| stampeder | but now I am curious. | 16:17 |
| sgw | did you change the boot cmdline? | 16:20 |
| *** yaawang_ has joined #starlingx | 16:32 | |
| *** yaawang has quit IRC | 16:33 | |
| stampeder | No, instead I changed the forced usb boot in my server's bios. I am now at the controller-0 login for Release 20.01. I want to thank you both for your help in this. Sometimes one needs another perspective. I have some issues with the Kubernetes GUI not allowing my system-admin rights properly, but I'll save that for later as the whole point was to | 16:39 |
| stampeder | check another image to see if it would cure the Kube Gui problem. | 16:39 |
| sgw | Are you seeing just a bunch of gibberish on the top left? | 16:40 |
| sgw | I think you need a new version of the k8s gui, I don't have the details right at hand. | 16:41 |
| sgw | stampeder: here are the details | 16:44 |
| sgw | The compatible version of dashboard to kubernetes 1.16 is v2.0.0-rc3, and the latest helm chart of dashboard is 1.10.1. | 16:44 |
| sgw | Which means you need to setup dashboard by a raw yaml [1], now locates in your system ~/dashboard-v2.0.0-rc3.yaml. The http method is only valid through the localhost access now, apparently, our access http://oam-ip:30000 is no valid now.The new dashboard release has fixed the cert auto-generation issue, so it is ok to access dashboard via https://oam-ip:30000But Chrome/Chromium is not able to access a self-signed website, so that | 16:44 |
| sgw | you need to access dashboard by firefox and add exception.Now login with the admin-user token should work. | 16:44 |
| sgw | stampeder: I have not dealt with this directly, so I am forwarding what I have gotten. | 16:45 |
| jsomervi1 | sgw thanks Saul, BTW I cringe when I see scripts named liked dl-tarball.sh where the package type shouldn't matter ie. tarball, rpm, cpio archive, etc. who cares? | 16:51 |
| stampeder | Argghhhh!!! I've run into this before with that Chrome issue when working with Openstack. We just quite using Chrome because of their non standard security patches. However, I did try using Firefox and it gave the same results as Chrome. Since I am using the iso images for my builds do you happen to know which dated iso had the correct version fo | 16:51 |
| stampeder | the dashboard? | 16:51 |
| sgw | jsomervi1: yeah, there is a lot of script names, .lst file names that I cringe at all the time. | 16:51 |
| sgw | It get's an initial name and gets stuck with that as it's usage expands. | 16:52 |
| jsomervi1 | sgw I just hope it doesn't get worse with all the working from home and potential day drinking.... | 16:52 |
| sgw | stampeder: it might be the dashboard version then | 16:53 |
| sgw | jsomervi1: that's reserved for friday afternoons, I have my homebrew and no one to share it with! | 16:53 |
| stampeder | sgw: I really think you are correct. | 16:53 |
| sgw | stampeder: on the day drinking or dash board version ;-) | 16:54 |
| stampeder | sqw: Both. Sometimes this stuff can drive you to drink. Only problem is I have a wheat allergy and I love dark beers......'=( | 16:55 |
| sgw | dpenney__: regarding the keystone puppet changes, are you good with the puppet changes, not something I can fully understand yet. Maybe another +1 from a puppet person (who?) | 16:57 |
| dpenney__ | sgw: my concern there was around whether the public endpoint is a security concern... I can ping Matt to see if he can take a look, too | 17:01 |
| sgw | Thanks that would be great. | 17:02 |
| sgw | dpenney__: slittle1: the dl_tarball patches for the kernel-rt srpm creation in tools and kernel ready for review. | 17:14 |
| *** sgw has quit IRC | 18:48 | |
| *** sgw has joined #starlingx | 20:16 | |
| * sgw back my comcast line went down! First time in all the years here | 20:17 | |
| *** stampeder has quit IRC | 21:24 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!