[05:15] <fluvvell> our lad installed desktop on a server by mistake, without a reinstall can we remove the package that requires a login to X before boot is complete? Crossgrade to ubuntu-server by package?
[05:16] <fluvvell> really don't want to log into a graphic login before we can ssh in
[06:52] <lordievader> Good morning
[06:53] <lordievader> fluvvell: Those are separate services. However, removing X libraries will remove most of the desktop, as those packages depend typically on X.
[10:59] <rafaeldtinoco> morning o/
[10:59] <lotuspsychje> morning rafaeldtinoco
[10:59] <rafaeldtinoco> o/
[14:03] <jamespage> cpaelzer: I'm seeing libipsec-mb0 being pulled into the depends for OVS in eoan via librte-pmd-aesni-mb18.11
[14:03] <jamespage> is that something you are aware of and was it intentional?
[14:03] <jamespage> i.e. do we need to MIR tat
[15:24] <lordcirth> Having a weird issue with netplan on 18.04 - I applied a config that I've used on other servers, vlan interfaces tied to a physical, bridges tied to vlans for lxc containers. The bridges don't start on boot
[15:24] <lordcirth> Boot waits for 2 minutes, 'brctl add' can add the bridges fine after boot. Must be some kind of race.
[15:31] <lordcirth> Ahah, I had configured the physical interface that the vlans tie to, in both the vlan file and the main file.
[15:32] <lordcirth> It would help for netplan to actually say something, though...
[15:32] <cyphermox> lordcirth: some things are hard to detect
[15:32] <cyphermox> could you open a bug and include both yaml files so we can add that?
[15:33] <lordcirth> cyphermox, sure. Ubuntu launchpad bug?
[15:33] <cyphermox> yes, please
[15:45] <Akuw> i am getting this  “Network configuration timed out: please verify settings”
[15:45] <Akuw> i have 4 ethernet cards but get that message, only one is connected to cable
[15:46] <Ussat> ok, verify the settings
[15:47] <Akuw> i am using default DHCP
[15:47] <Akuw> all cards are in the list
[15:47] <Akuw> but got that error
[15:48] <Akuw> Broadcom Limited / Netextreme ii BCM5709 Gigabit Ethernet
[15:48] <Akuw> is enp4s0f0
[15:48] <Akuw> link is blinking
[15:52] <Akuw> dhclient enp4s0f0 stay waiting
[15:52] <Akuw> cant get any ip
[15:58] <tomreyn> Akuw: are you using the ubuntu *live* server installer, or the (alternative) server installer? for which ubuntu version?
[15:58] <Akuw> ubuntu-18.04.2-live-server-amd64.iso
[16:01] <Akuw> ping to 127.0.0.1 gives response
[16:01] <Akuw> then card is ok
[16:02] <tomreyn> Akuw: can you ping the gateway, though?
[16:02] <Akuw> no
[16:02] <Akuw> i cant get DHCP
[16:03] <Akuw> ubuntu server should get automatically because i selected DHCP
[16:03] <Akuw> for IPv4 and IPV6
[16:13] <tomreyn> Akuw: yes, nrmally this is what would be expected to happen. and i would think this NIC should be supported, too. you could check the logs which should be somewhere in /var/log
[16:13] <tomreyn> Akuw: maybe /var/log/installer
[16:15] <Akuw> i should disable PXE Boot from this card in BIOS or doesnt matter?
[16:15] <tomreyn> Akuw: here's a log file of a successful NIC configuration on 18.04.2 server-live installer:  https://termbin.com/0yo6
[16:16] <tomreyn> Akuw: pxe boot should no longer matter after you already booted the installer.
[16:17] <Akuw> ok
[16:17] <tomreyn> Akuw: have you tried to cross check with a desktop installer / live system?
[16:17] <Akuw> no
[16:18] <tomreyn> i.e. it'd be good to test whether this is really an issue with the server-installer or whether the DHCP configuration is just incorrect.
[16:18] <Akuw> let me see log
[16:18] <Akuw> this server is old so should be work
[16:19] <tomreyn> what i'm suggesting is that the network configuration data transferred via dhcp might be wrong so it cannot actually connect anywhere
[16:19] <tomreyn> this log is /var/log/installer/subiquity-debug.log
[16:19] <tomreyn> (in case you'd like to compare to yours)
[16:19] <Akuw> no
[16:19] <Akuw> i tested from another computer
[16:20] <Akuw> so netwrok is ok
[16:20] <Glorfindel> hi all, I need to install mysql 5.2 on ubuntu 18.04. what is the easiest way to go about doing this? I'm hoping to avoid building from source
[16:21] <Akuw> i will test with desktop
[16:21] <Akuw> wait
[16:21] <Akuw> letme create usb boot
[16:22] <Glorfindel> mysql-server 5.2, specifically
[16:22] <tomreyn> Glorfindel: why do you need this old version when you can use a newer one? and why do you depend on installing it on 18.04?
[16:23] <Glorfindel> tomreyn: it's the newest version supported by some oldish software I'm trying to get running
[16:27] <tomreyn> Glorfindel: since the oldest mysql server verison ubuntu supports at this time is mysql 5.7 (5.5 via !ESM) you'd indeed need to find another source for installing it. some of the mysql community distriubtions do provide their own versioned apt repositories, but i doubt they still maintain such an old release.
[16:27] <tomreyn> i'd rather suggest you choose a maintained software.
[16:31] <Glorfindel> the software in question is partkeepr, and from the research I did before deciding to test it there isn't really anything else availible. I was hoping to use it for inventory tracking
[16:31] <lordcirth> Glorfindel, there are many options for inventory software
[16:33] <Akuw> hi
[16:33] <Akuw> i just did a test
[16:33] <Akuw> from another computer it was possible to get IP using DHCP
[16:33] <Akuw> using same cable
[16:34] <Glorfindel> lordcirth: well, within the feature set I was looking for, I guess. do you have any alternatives to suggest that maybe didnt' show up when I searched?
[16:34] <Akuw> so... wat could be the problem?
[16:35] <Akuw> network card?
[16:35] <Akuw> driver?
[16:35] <lordcirth> Glorfindel, not off-hand. But I would not consider anything that requires mysql 5.2.
[16:36] <tarpman> Glorfindel: where do you see that partkeepr requires mysql 5.2? I see plenty of people on their github who appear to be running it on recent-ish ubuntu
[16:36] <tarpman> Glorfindel: https://wiki.partkeepr.org/wiki/PartKeepr_on_Debian_%22Stretch%22 even talks about running it on mariadb 10
[16:37] <Akuw> tomreyn: what do you think could be the problem, Ubuntu install is OK, Cable is OK
[16:37] <Glorfindel> tarpman: it's currently not running and when I asked in #partkeepr I was told 5.2 is the most recent mysql version it will run on
[16:37] <Akuw> the only thing could be network card
[16:40] <Glorfindel> although they didn't sound like they remembered for sure either... meh. maybe I'll try finding an alternative... I've already spent around 8 hours trying to get this one working
[16:52] <tomreyn> Glorfindel: what are your requirements in an inventory tracking system?
[17:01] <Glorfindel> tomreyn: barcode scanner support; notifications of low stock; automatic upc lookup would be nice, but not necessary. honestly I haven't really seen much of anything as far as inventory goes...
[17:04] <tomreyn> Glorfindel: this doesn't sound special at all, i guess most ERP solutions will cover this.
[17:05] <tomreyn> https://en.wikipedia.org/wiki/Inventory_management_software#See_also
[17:05] <tomreyn> wikipedia usually has lists of commonly used softwares for specific tasks, too, have a look
[17:10] <Glorfindel> tomreyn: I don't see any lists of commonly used software :/
[17:12] <tomreyn> https://en.wikipedia.org/wiki/List_of_ERP_software_packages
[17:13] <tomreyn> ERP may be much more than you want / need though, i guess you just want inventory management
[17:13] <tomreyn> or warehouse mgmt
[17:18] <tomreyn> or supply chain mgmt. but i agree there dont seem to be a lot of comparisons for those online, which probably suggests those are not the right search terms, yet.
[17:20] <tomreyn> Glorfindel: https://ofbiz.apache.org/ might be an option (this is not a recommendation)
[17:25] <tomreyn> maybe https://github.com/odoo/odoo also
[17:49] <tomreyn> i'm looking at http://cdimage.ubuntu.com/ubuntu-server/bionic/daily-live/20190802/MD5SUMS - are arm64 + ppc64el going to be built (and released) as well?
[17:50] <tomreyn> ahem, actually thes are on the same directory, sorry.
[17:50] <tomreyn> they're ust not listed at http://iso.qa.ubuntu.com/qatracker/milestones/405/builds which makes me wonder what their status is
[18:40] <Glorfindel> tomreyn: hmm, I'll keep those in mind. thanks. right now I'm thinking inventoria from nch, seems to have everything I'll need and runs on windows, which is a plus. Not open source or free, but a lifetime license costs less than some monthly subscriptions from other companies
[18:43] <DammitJim> I need to add more space to a couple of LVs; however my PVs are on a software raid
[18:43] <DammitJim> do you guys know how I can achieve this? This is in a vmware environmen
[18:44] <tomreyn> Glorfindel: whatever works for you ;) we're way beyong the scop of #ubuntu-server now, thoough - let's continue in #ubuntu-offtopic or -discuss if there's more to talk about it.
[18:45] <tomreyn> DammitJim: so you have no unallocated space left in any VGs?
[18:45] <DammitJim> correct
[18:45] <DammitJim> I've added more space to a VG
[18:45] <DammitJim> but I've never done it to a VG that is backed by a software raid
[18:55] <tomreyn> DammitJim: well the software has a fixed size, so the PV on top of it has a fixed maximum size. you mentioned multiple PVs though, so it's hard to imagine what you are really dealing with there.
[18:56] <tomreyn> sharing some command output (on a !pastebin - consider the !pastebinit command) might help
[18:56] <tomreyn> !pastebin
[19:19] <DammitJim> where did I mention multiple PVs?
[19:20] <tomreyn> mwhudson: in case i can bug you with htis currently (feel free to point me elsewhere / not respond) - i'm trying today's  bionic-live-server-amd64.iso from 18.04.3 qa-testing (a8dff1ad76cf8b8424c9695a2ae13b9a) in virtualbox (6.0.10 r132072) with 4 NICs (3 of which have no link) and run into e1000 "Tx Unit Hang" errors: https://termbin.com/3sho https://i.imgur.com/NXy2ogc.png
[19:21] <tomreyn> also a netdev watchdog oops as a result of those
[19:21] <tomreyn> should i report this somewhere / omehow? (i'm not into QA testing, yet)
[19:22] <tomreyn> DammitJim: "however my PVs are on a software raid" (plural)
[19:25] <DammitJim> oh, thanks for pointing that out
[19:25] <DammitJim> it's 1 PV
[19:29] <tomreyn> DammitJim: so you have a couple of (virtual) HDDs, you have partitions on those, and on one of those parititons you have an mdadm managed software raid? correct? and which raid level is it, how many raid devices?
[19:29] <tomreyn> cat /proc/mdstat    also works
[19:32] <tomreyn> DammitJim: and then on top of this RAID array you have a logical volume meanager 2.x PV, which is assigned to a VG, and all allocatable space has been allocated. now what is your strategy for increasing allocatable space within this volum group?
[19:39] <DammitJim> raid 1
[19:40] <DammitJim> I thought I could increase the virtual hard drives that are part of this raid array
[19:40] <DammitJim> so, I think lsblk shows something like sdc -> sdc1 -> md0 and sdb -> sdb1 -> md0
[19:40] <tomreyn> DammitJim: as discussed above, i assume your raid member devices as not full disks but partitons.
[19:40] <DammitJim> I'm picturing in my head that we would fdisk sdc and sdb with new sdc2 and sdb2
[19:40] <tomreyn> right
[19:40] <DammitJim> and add them to the raid array
[19:41] <DammitJim> yes, not full disks
[19:41] <DammitJim> but I don't know what commands I would run to add the new partitions to the raid array so they are mirrored
[19:41] <tds> DammitJim: can you pastebin the exact output of lsblk?
[19:41] <DammitJim> sure
[19:41] <tds> also, you mentioned VMs are involved - to confirm, is this inside the vm, on the host, or you're tweaking config on both, or what?
[19:41] <TJ-> DammitJim: is the space on sdb/sdc after sdb1/sdc1 currently unallocated?
[19:42] <tomreyn> you could repartition the existing disks to add new partitions and add those as new raid devices to a raid 10. but that's a lot more complexity that i'd recommend.
[19:42] <DammitJim> https://pastebin.com/97nfaH5Q
[19:42] <DammitJim> the existing disks are maxed out
[19:43] <Ussat> whych hypervisor ?
[19:43] <DammitJim> the VM is the system I want to add more storage to... I have plenty storage in the datastore - vmware
[19:43] <Ussat> Just add a physical disk, then ad that to the LV, then expand it
[19:44] <tomreyn> DammitJim: wait, do you have a mirror raid spun across different partitions on sda?
[19:44] <DammitJim> Ussat, I need to add it to the raid array first
[19:44] <DammitJim> I don't know how to do that
[19:44] <Ussat> or, use SSM
[19:44] <DammitJim> tomreyn, don't worry about sda... that's the non-raided section (that's why I originally said PVs)
[19:44] <DammitJim> sdb and sdc is where I need more room
[19:45] <Ussat> no you dont....the origional disk is in the array right ?
[19:45] <tomreyn> ok i'll try to ignore sda
[19:45] <Ussat> just add the space
[19:46] <DammitJim> no, the original disk is NOT in the array (this is all virtual and I'm about to expand the virtual hard drive by doing an "edit VM")
[19:47] <Ussat> wait I am confused here.....
[19:47] <Ussat> ...
[19:47] <Ussat> editing the VM like that is not reccomended
[19:48] <Ussat> if its a VM, where is this raid array you speak of ?
[19:48] <TJ-> DammitJim: so sdb/sdc are virtual disks. It is possible therefore to increase the space allocated to those from the hypervisor and in the guest re-write the partition tables so that sdb1/sdc1 now cover the larger 'disks' using something like "mdadm --grow --size=X ..." and then all you need is a "pvresize /dev/sdb1; pvresize /dev/sdc1"
[19:49] <DammitJim> TJ-, the problem is that the drives are already 2TB and fdisk doesn't like anything larger than that
[19:49] <DammitJim> so, I was thinking I'd have to do a new sdb2 and sdc2
[19:49] <DammitJim> am I wrong?
[19:49] <TJ-> DammitJim: 'rewrite the partition tables' would require using something like sfdisk (MSDOS label) or sgdisk (GPT label)
[19:49] <Ussat> dont use fdisk, use SSM
[19:50] <TJ-> DammitJim: depending on which sector partition #1 starts in sdb/sdc you can convert to GPT easily
[19:51] <TJ-> DammitJim: I'd guess partition #1 start at sector 2048, in which case a conversion to GPT is entirely possible, since it only needs sectors 1-33 (plus it will but a secondary table at end-of-disk )
[19:51] <Ussat> http://manpages.ubuntu.com/manpages/xenial/man8/ssm.8.html
[19:53] <tds> "You can not create or manage MD volumes or pools, but it will be  extended  in the future." sounds like it'll be pretty useless for this situation, for the md resize at least
[19:53] <tds> personally i'd much rather do it by hand as described by TJ- than having some magic tool do it all anyway
[19:54] <TJ-> Ussat: does ssm enable taking over an existing config?
[19:54] <Ussat> yes
[19:54] <Ussat> Its godlike, have been useing it for a long time
[19:56] <tomreyn> your god has recently deceased, though, according to packages.ubuntu.com
[19:56] <tomreyn> https://packages.ubuntu.com/search?keywords=system-storage-manager
[19:58] <TJ-> looks like there were only 2 releases https://launchpad.net/ubuntu/+source/system-storage-manager
[19:58] <tomreyn> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=849671
[19:59] <tomreyn> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=845517
[20:06] <DammitJim> what now?
[20:07] <tomreyn> TJ-'s suggestion of using sfdisk (MSDOS label) or sgdisk (GPT label) seemed reasonable to me.
[20:25] <DammitJim> sgdisk, got it
[20:26] <DammitJim> then how do I add the partitions to the array?
[20:28] <Ussat> https://www.digitalocean.com/community/tutorials/how-to-manage-raid-arrays-with-mdadm-on-ubuntu-16-04
[20:28] <Ussat> http://www.ducea.com/2009/03/08/mdadm-cheat-sheet/
[22:19] <Numbers23> Hi all anyone have experience with running Hbase in pseudo distributed mode? I can establish a shell session but when I go to try create a new table I get the error: "ERROR: KeeperErrorCode = NoNode for /hbase/master"
[22:19] <sarnold> any chance you get a more detailed error message in the hbase logs?
[22:21] <sarnold> man stackoverflow's a wasteland on this one.. heh
[22:22] <sarnold> the most useful answer I've found yet is deleted; it involes "start zookeeper, master, regionserver"
[22:23] <sarnold> Numbers23: aha, there's something here that looks tolerable :) https://hub.packtpub.com/5-mistake-developers-make-when-working-hbase/  "When thezookeeper.znode.parentconfig value on the client side doesn’t match the one for your cluster"\
[22:23] <sarnold> "One possible scenario is that hbase-site.xml is not on the classpath of the client application.The default value for zookeeper.znode.parent doesn’t match the actual one for your cluster. When you get hbase-site.xml onto the classpath, the problem should be gone."
[22:25] <Numbers23> Thanks for your replies sarnold, not sure if this helps "bin/hbase shell
[22:25] <Numbers23> 2019-08-02 23:05:15,752 WARN  [main] util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
[22:25] <Numbers23> HBase Shell
[22:25] <Numbers23> Use "help" to get list of supported commands.
[22:25] <Numbers23> Use "exit" to quit this interactive shell.
[22:25] <Numbers23> For Reference, please visit: http://hbase.apache.org/2.0/book.html#shell
[22:25] <Numbers23> Version 2.2.0, rUnknown, Tue Jun 11 04:30:30 UTC 2019
[22:25] <Numbers23> Took 0.0143 seconds
[22:25] <Numbers23> hbase(main):001:0> create 'usertable', 'cf1'
[22:25] <Numbers23> ERROR: KeeperErrorCode = NoNode for /hbase/master"
[22:27] <Numbers23> hbase-site.xml is saved under usr/local/hbase/conf
[22:28] <sarnold> Numbers23: and is it identical for all the servers? is it necessary for the client? if so, does the client have an identical copy too?
[22:31] <Numbers23> sorry I'm new to all of this not sure I understand your questions. Trying to create a table in Hbase and test its performance through YCSB vs Cassandra's performance
[22:31] <Numbers23> (it's a college project)
[22:32] <sarnold> Numbers23: it could be my fault, I'd only read about hbase for a few days a few years ago
[22:33] <sarnold> Numbers23: how many servers do you have running?
[22:33] <sarnold> Numbers23: do they all have identical configurations?
[22:38] <Numbers23> In the /etc/host file do you mean? I have the local host IP and then my Virtual Box IP.
[22:38] <sarnold> Numbers23: how many hbase servers are you running?
[22:42] <Numbers23> I'm sorry I don't know how to check that :/
[22:43] <sarnold> Numbers23: you may need to ask whoever set it up
[22:43] <Numbers23> Yes more research needed on my part. Thanks for taking the time to answer me though :)
[22:45] <sarnold> Numbers23: fwiw I really enjoyed the Seven Databases in Seven Weeks book; it's a nice fast introduction to a handful of different databases
[22:45] <sarnold> Numbers23: it's a coupleyears old at this point, but I expect the hbase introduction on it might be close enough to what you've got now, you could probably be up and running with your own server pretty quick, while you're waiting to hear back from whoever runs yours :)
[22:51] <Numbers23> haha true that, I appreciate the recommendation. Hopefully I can get through it in 7 days rather than 7 weeks!
[22:52] <sarnold> hehe yeah, each chapter is designed for three days, but, eh, you know, go at your own pace :)
[23:07] <Greyztar> hello,im running a cron job every minute,ive turned off logging in cron using the extra opts variable to avoid spam in syslog but im still seeing cron entries in syslog related to pam_unix session opening and closing everytime that cron job is running,can i turn these messages off somehow?
[23:13] <sarnold> Greyztar: I think you can, but it's a bit off the usual path.. so you might introduce other oddities later
[23:13] <sarnold> Greyztar: first I have to suggest that you change your application a bit, so you run your thing once, at boot, and then your application sleep for N seconds between executions, or something similar
[23:14] <sarnold> Greyztar: but if you still want to continue.. try running pam-auth-update and see if it makes it easy for you to remove the *session* component of pam_unix from the cron service. (I'm skeptical, but if it can do it, that'd be best)
[23:15] <Greyztar> sarnold: ahh ok ill look into that then thanks!
[23:15] <sarnold> Greyztar: if that doesn't offer it, then you'll have a bit more work on your hands. it's not terrible, but it's not particularly future-proof either
[23:16] <sarnold> Greyztar: you'd inline everything from /etc/pam.d/common-session into /etc/pam.d/cron and remove the pam_unix *session* portion. just session. only session. :)
[23:18] <Greyztar> sarnold: interesting ill have to try that out,will have to be tomorrow though its 01:18 here so im off now,thanks for help though and have a good day/night =)
[23:18] <sarnold> gnight Greyztar, have fun :)
[23:19] <sarnold> (and yeah, PAM changes at 1am.. risky :)
[23:19] <Greyztar> cheers (,")