Back to Home / #linode / 2019 / 12 / Prev Day | Next Day
#linode IRC Logs for 2019-12-30

---Logopened Mon Dec 30 00:00:46 2019
01:25-!-chatpeople [~oftc-webi@119.148.4.18] has joined #linode
01:25-!-chatpeople is "OFTC WebIRC Client" on #linode
01:26<chatpeople>hello
01:26<chatpeople>anybody there for help
01:28<millisa>!ask
01:28<linbot>If you have a question, feel free to just ask it -- someone's always willing to help. If you don't get a response right away, be patient! You may want to read http://alexfornuto.com/how-to-ask-for-help-on-irc/
01:31<chatpeople>i want to increase my linode disk size
01:31<chatpeople>how to do that
01:31<chatpeople>@millisa
01:31<millisa>Disk allocation is tied to the plan size.
01:31<millisa>There's also the block storage that can be added on
01:32<millisa>Info on block storage is here: https://www.linode.com/docs/platform/block-storage/how-to-use-block-storage-with-your-linode/
01:32<millisa>If you haven't allocated all the disk in your plan, you can increase the disk following this guide: https://www.linode.com/docs/quick-answers/linode-platform/resize-a-linode-disk/
01:32<chatpeople>you mean the volume? option?
01:33<millisa>Info on resizing a linode to one that has a larger disk allocation would be here: https://www.linode.com/docs/platform/disk-images/resizing-a-linode/
01:36<chatpeople>you are telling i have to change my plan right?
01:36<chatpeople>withotut changing the plan, i can not increase disk only
01:36<millisa>If you have allocated all the disk and you don't want to add on block storage, yes?
01:37<chatpeople>i want to stick with current plan but want to increase the disk size only
01:40<chatpeople>Would you like the disk CentOS 7 Disk to be automatically scaled with this Linode's new size? We recommend you keep this option enabled.
01:40<chatpeople>Additional verification is required to add this service. Please open a support ticket.
01:41<chatpeople>when i am going to resizing the plan it is showing Additional verification is required to add this service. Please open a support ticket.
01:41<millisa>Ok
01:44<chatpeople>what shall i do
01:45<millisa>I'm assuming you'd open a support ticket
01:47-!-thiras [~thiras@195.174.212.70] has quit [Ping timeout: 480 seconds]
01:50<chatpeople>hi
01:51<millisa>er, hi
01:51<chatpeople>can you help
01:51<chatpeople>are you from support team
01:52<millisa>!ops
01:52<linbot>Users with ops are employees of Linode, and know what they're talking about. The rest of us are the ever-so-helpful(?) community. Official Linode contact information: https://www.linode.com/contact
01:53<chatpeople>hi millisa
01:53<millisa>hi, again
01:54<chatpeople>you from linode support team
01:55<millisa>Users with ops are employees. The rest of us are users.
01:57<chatpeople>understood
01:57<chatpeople>can you tell me "why is is showing"
01:57<chatpeople>"Additional verification is required to add this service. Please open a support ticket."
01:57<chatpeople>why it is showing when i am going to resize
01:58<millisa>Probably because they need extra verification. I'd guess you have a new account and they want to be sure you aren't creating a large bill that you won't expect
01:59<chatpeople>i got it
02:02<chatpeople>nothing working
02:02<chatpeople>Linode has allocated more disk than the new service plan allows. Delete or resize disks smaller.
02:02<chatpeople>i think most of the given information are fake
02:02<chatpeople>i was trying to go to lower plan
02:02<chatpeople>i did not consume much spaces but it is showing different
02:02<millisa>Your disk has to be small enough to fit into the smaller plan
02:08<chatpeople>sometimes it is showing Linode applications-master has been booted by the Lassie watchdog
02:08<chatpeople>what is lassie watchdog
02:08<chatpeople>will it reboot the server
02:08<millisa>Linode Autonomous System Shutdown Intelligent rEbooter
02:09<millisa>here's a 12 year old article on it. https://www.linode.com/2007/10/26/lassie-the-shutdown-watchdog/
02:09<chatpeople>haha
02:09<chatpeople>just want to know
02:09<chatpeople>will it reboot the server or
02:11<millisa>they briefly mention it here: https://www.linode.com/docs/uptime/monitoring-and-maintaining-your-server/#configure-shutdown-watchdog
02:12<chatpeople>i have one more question
02:12<chatpeople>how to recover the server if it crashes
02:12<chatpeople>basically we are very new into this data center
02:12<chatpeople>still learning many things
02:13<chatpeople>particularly previous all data and how much time it required to shift to new server with all previous data
02:13<chatpeople>"particularly previous all data and how much time it required to shift to new server with all previous data"
02:13<millisa>there's the lish console if you need to see what it looks like from the console. there's also a rescue mode you can put the system in. Details here: https://www.linode.com/docs/troubleshooting/rescue-and-rebuild/
02:14<millisa>Details on lish are here: https://www.linode.com/docs/platform/manager/using-the-linode-shell-lish/
02:15<millisa>consider buying their backup addon - with that you can restore one of the backups into a linode if necessary
02:16<chatpeople>yes taking backup for each server
02:19<chatpeople>can you tell me
02:19<chatpeople>how to deal with backup
02:19<chatpeople>any tutorial
02:19<chatpeople>i mean when and how backup will help me at what condition
02:19<millisa>https://www.linode.com/docs/platform/disk-images/linode-backup-service/
02:41<chatpeople>one help
02:54-!-AugustusCaesar24 [~AugustusC@99-190-112-116.lightspeed.irvnca.sbcglobal.net] has joined #linode
02:54-!-AugustusCaesar24 is "Augustus Caesar" on #linode
03:26-!-chatpeople [~oftc-webi@119.148.4.18] has quit [Remote host closed the connection]
03:57-!-AugustusCaesar24 [~AugustusC@99-190-112-116.lightspeed.irvnca.sbcglobal.net] has quit [Quit: Going offline, see ya! (www.adiirc.com)]
07:05-!-Juma [~amir@77.139.185.199] has joined #linode
07:05-!-Juma is "Amir Uri" on #linode
07:22-!-fstd_ [~fstd@xdsl-87-78-190-112.nc.de] has joined #linode
07:22-!-fstd_ is "fstd" on #oftc #linode #debian #kernelnewbies
07:30-!-fstd [~fstd@xdsl-87-78-140-227.nc.de] has quit [Ping timeout: 480 seconds]
07:51<linbot>New news from community: How do I fix system issues after upgrade to Centos 7.7 <https://www.linode.com/community/questions/19262>
08:03-!-pawan [~oftc-webi@51.235.118.48] has joined #linode
08:03-!-pawan is "OFTC WebIRC Client" on #linode
08:04<pawan>Hi! My domain is msh.com.sa. My mail server on another server. I want to route the mails to that server. Could you please help .
08:06<pawan>??
08:07<grawity>just add a MX record that points to the new server's domain name
08:07<grawity>oh yeah and update the SPF record (the TXT "v=spf1" one) with the new server's addresses
08:08<pawan>And what abt A record?
08:09<nate>Question to fellow PCI-DSS people, if you're doing any sort of card transactions (storing them or not even), user accounts are supposed to have passwords correct and follow the minimum PCI-DSS guidelines right?
08:09<grawity>the only A/AAAA records that are used for email delivery are those named by MX records
08:09<nate>Pretty sure that hasn't changed at all with recent PCI-DSS specifications but I haven't really reviewed them in a while
08:09<grawity>so if you have MX pointing to "mail.msh.com.sa.", then only the mail.msh.com.sa A/AAAA records are used
08:11<pawan>Got it!! Thanks a lot.
08:13-!-pawan [~oftc-webi@51.235.118.48] has quit [Quit: Page closed]
08:37-!-Juma [~amir@77.139.185.199] has quit [Ping timeout: 480 seconds]
08:50-!-Juma [~amir@77.139.185.199] has joined #linode
08:50-!-Juma is "Amir Uri" on #linode
08:56-!-eyepulp [~eyepulp@107.152.3.83] has joined #linode
08:56-!-eyepulp is "eyepulp" on #linode
09:00-!-Juma [~amir@77.139.185.199] has quit [Ping timeout: 480 seconds]
09:46-!-w_ [~w@24.133.136.184] has joined #linode
09:46-!-w_ is "Alper Akyıldız" on #linode
09:47-!-w_ [~w@24.133.136.184] has left #linode []
10:45<linbot>New news from community: Is it appropriate to use Linode object storage for web app back ends? <https://www.linode.com/community/questions/19263>
10:49-!-DrJ [~asdf@li1303-21.members.linode.com] has joined #linode
10:49-!-DrJ is "DrJ" on #linode
10:57<linbot>New news from community: NodeBalancer support for PROXY protocol <https://www.linode.com/community/questions/19264>
11:45-!-mwp [sid206092@2001:67c:2f08:5::3:250c] has joined #linode
11:45-!-mwp is "Mike Pastore" on #linode
11:49<mwp>Hi folks, I have a Linode with two dedicated CPUs, and I'd like to set the default CPU affinity to vcpu #1 so I can manually put some other processes on vcpu #2. Ubuntu Server 18.04 LTS with the current 64-bit Linode kernel. I set CPUAffinity=1 in /etc/systemd/system.conf, executed `systemctl daemon-reload`, and rebooted, but SystemD is still putting some processes on vcpu #2. Any suggestions? Thank you in advance.
12:22<linbot>New news from community: Server not responding after reboot. <https://www.linode.com/community/questions/19266> || Network Transfer Question? <https://www.linode.com/community/questions/19265>
12:28-!-shubhamkjha [~shubhamkj@47.9.212.146] has joined #linode
12:28-!-shubhamkjha is "ShubhamKJha" on #linode
12:32<shubhamkjha>Hi
12:32<millisa>Greetings
12:33<shubhamkjha>I am trying to configure irc for the first time
12:33<shubhamkjha>:)
12:33<millisa>It would appear you are having some success
12:34<shubhamkjha>true
12:34<shubhamkjha>looks too much geeky though
12:35-!-shubhamkjha [~shubhamkj@47.9.212.146] has left #linode []
12:36<millisa>Too Much Geeky since 1988
12:37<Peng_>D:
12:51<@rgerke>mwp: Sorry it took me a bit to respond, but I do not have an answer for you. I've brought this up to the Support team to see if we can come up with a suggestion for you.
12:51<mwp>Great! Thanks @rgerke
12:51<mwp>I might have to set isolcpus on the kernel command line but it doesn't look like that's possible with the Linode kernel
12:52<mwp>and I'm loathe to run and maintain grub and a distro kernel
12:53<@rgerke>mwp: Hang tight for a bit - I'll have something for you as soon as I can!
12:53<mwp>tks
12:56-!-Shentino_ [~desktop@96-41-208-125.dhcp.elbg.wa.charter.com] has joined #linode
12:56-!-Shentino_ is "realname" on #qemu #mm #linode #tux3
12:58-!-Shentino [~desktop@96-41-208-125.dhcp.elbg.wa.charter.com] has quit [Ping timeout: 480 seconds]
13:01<dwfreed>mwp: pretty sure that value is 0-indexed, so you're telling systemd to put all processes on the second core, and it's doing exactly what you asked; look at /proc/<pid>/status for the Cpus_allowed_list value
13:01<mwp>ooooh
13:03<@jackley>hmmm I think that might be right?
13:03<@jackley>http://man7.org/linux/man-pages/man2/sched_setaffinity.2.html
13:03<dwfreed>jackley: look at the example
13:03<dwfreed>it gives 0-based processor IDs
13:04<dwfreed>also references /proc/cpuinfo, which is also 0-based
13:04<@jackley>ya was just looking at that!
13:04<mwp>hrrm I think the syscall is zero indexed but the systemd parameter is one indexed
13:04<dwfreed>but, for fun, I can spin up a systemd service with CPUAffinity set
13:04<dwfreed>grep Cpus_allowed_list /proc/*/status | sort | uniq -c
13:04<@jackley>There are various ways of determining the number of CPUs available on the system, including: inspecting the contents of /proc/cpuinfo; using sysconf(3) to obtain the values of the _SC_NPROCESSORS_CONF and _SC_NPROCESSORS_ONLN parameters; and inspecting the list of CPU directories under /sys/devices/system/cpu/.
13:05<dwfreed>after you tell grep not to print the filename...
13:05<dwfreed>(-h)
13:05<@jackley>`cat /proc/cpuinfo` lists my first CPU as 0
13:07<mwp>it's definitely cpu0 and cpu1
13:08<dwfreed>note that some processes may adjust their affinity themselves
13:08<dwfreed>nginx can do this, for example
13:08<@jackley>https://www.freedesktop.org/software/systemd/man/systemd-system.conf.html
13:08<@jackley>ndividual services may override the CPU affinity for their processes with the CPUAffinity= setting in unit files, see systemd.exec(5).
13:08<@jackley>ya – what dwfreed said
13:09<mwp>yup aware of that bit
13:13<dwfreed>CPUAffinity= is definitely 0-indexed as well
13:14<dwfreed>just tested it with a simple service
13:14<@jackley>just noodling around on a dedicated Linode that I set up to test – setting CPUAffinity=0 seems to have done it for me.
13:14<@jackley>everything is on CPU 0
13:14<dwfreed>:)
13:14<@jackley>!point dwfreed
13:14<linbot>jackley: Point given to dwfreed. (90) (Biggest fan: mcintosh, total: 17)
13:15<dwfreed>grep -h Cpus_allowed_list /proc/*/task/*/status | sort | uniq -c
13:15<dwfreed>^ to see the count of all the different affinity lists; it is a per-thread property
13:20<mwp>so this is definitely moving me in the right direction
13:21<mwp>the problem now is that 91 tasks show Cpus_allowed_list 0-1
13:21<mwp>these are the processes not started by SystemD, so the CPUAffinity= directive does not apply
13:21<mwp>and the solution (AFAICS) is to set isolcpus on the command line
13:22<dwfreed>yes
13:22<dwfreed>that'll set the default affinity, and then individual processes can override it
13:22<dwfreed>should take that 91 down quite a bit, at the very least
13:22<mwp>so is there any way to do that on the linode kernel? or do I have to install grub and a distro kernel?
13:23<dwfreed>you'd have to use distro kernel
13:23<dwfreed>(or custom)
13:23<mwp>gotcha
13:23<@jackley>tbh maintaining the distro kernel is just as easy as Linode's, once it's set up
13:23<mwp>okay thanks all... dwfreed jackley rgerke
13:23<mwp>much appreciated
13:23<@jackley>(the distro kernel is our default for new images)
13:23<@jackley>np!
13:23<dwfreed>you don't need to install grub as the bootloader, but you do need the pieces that generate the /boot/grub/grub.cfg
13:23<mwp>ahh gotcha
13:24<mwp>yeah I hear you about the distro kernel, there's just something oddly reassuring about using the linode kernel for everything
13:24<@jackley>fair enough
13:24<dwfreed>(you'd only need to install grub as a bootloader if you were using direct disk boot; there's the GRUB "kernel" option, which uses a Linode-provided GRUB to do the bootloader bits and just uses your /boot/grub/grub.cfg for the config)
13:24<@jackley>(if you needed a guide, here's one https://www.linode.com/docs/platform/how-to-change-your-linodes-kernel/)
13:25<mwp>also ubuntu hwe is still at 5.0 only
13:25<mwp>which would be the low maint sol'n
13:27-!-CodeMouse92 [~JasonMc92@00025241.user.oftc.net] has joined #linode
13:27-!-CodeMouse92 is "Jason C. McDonald" on #packaging #linode #c++
13:30<linbot>New news from community: k8s volume reattachment takes 7 minutes after node disconnect! <https://www.linode.com/community/questions/19267>
14:04<grawity>dwfreed: actually, are those pieces still needed anymore? IIRC nearly all of them were to do with Lish acting as a serial console, and that's kind of irrelevant when Glish exists
14:20-!-eyepulp [~eyepulp@107.152.3.83] has quit [Remote host closed the connection]
14:52<linbot>New news from community: Kubernetes Service type LoadBalancer does not get external IP <https://www.linode.com/community/questions/19268>
15:15-!-reiindear [~oftc-webi@2604:6000:cc41:4700:7cfe:839b:4b5b:cab5] has joined #linode
15:15-!-reiindear is "OFTC WebIRC Client" on #linode
15:17<reiindear>Hi there I'm interested in using Linode and wondering if anyone can tell me how hard/easy it is to upgrade a virtual machine to a higher level once I've set it up. Like can I press a button to move from a $10 a month server to a $20 a month one and keep everything I've already done or will I need to reinistall everything on the new server?
15:18<reiindear>Thank you for any info on this. I'm sure it's written somewhere on the site but I cant seem to find it in the plan information
15:19<@jackley>reiindear: you can press a few buttons and just move to a larger (or smaller) plan -> https://www.linode.com/docs/platform/disk-images/resizing-a-linode/
15:20<reiindear>jackley: thank you
15:20<@jackley>reiindear: you're welcome!
15:22<reiindear>unrelated but I've found a possible bug on the sales page https://www.linode.com/products/standard-linodes/. Underneath "Solutions to easily deploy and scale " there are a number of tabs " One-Click Apps StackScripts Distributions " and " Storage Networking Tools Services" which are not clickable or at the very least don't change the visible tab when clicked.
15:23<reiindear>I have tested this in firefox 66.0.3 (64-bit) and chrome 79.0.3945.88. firefox I enabled all scripts and chrome is a vanilla install
15:23<reiindear>Either way thank you all this is by far one of the best irc channels I've visited in a while
15:27<@jackley>hmmmm
15:27<@jackley>reiindear: meaning when you click on "Stackscripts", nothing happens? that's what I'm experiencing (Firefox 70)
15:27<@jackley>reiindear: and thank you! <3
15:29<reiindear>yes that's the behavior I experience as well
15:30<@jackley>seems like a problem. i'll pass it along, thanks for spotting & reporting that!
15:30<reiindear>you're welcome
15:31-!-reiindear [~oftc-webi@2604:6000:cc41:4700:7cfe:839b:4b5b:cab5] has quit [Quit: Page closed]
16:14<linbot>New news from community: What do I do if I see AVC denial errors? <https://www.linode.com/community/questions/19269>
16:20<linbot>New news from blog: 2019: A Big Year of Innovation for Linode and a Big Thank You to Our Customers <https://www.linode.com/2019/12/30/2019-a-year-in-review/>
16:23*Peng_ tempted to make many bad terms of service jokes
16:24<Peng_>https://www.juniper.net/us/en/company/case-studies-customer-success/linode/ <- Interesting link in the middle of that post.
16:24<Peng_>Wait, wasn't Linode a Cisco shop?
16:24<virtual>I thought so too?
16:25<virtual>but my knowledge is at least 4 years old. much can happen in that time.
16:27<Peng_>"Starting today, we’ve reduced our transfer overage price to $0.01." \o/
16:27<virtual>ooh, what was it?
16:28<virtual>oh, 0.02. 50%! :)
16:30<FluffyFoxeh>don't stretch yourselves too thin, now :)
16:34<virtual>this is a big discount, for some people, I am sure.
16:34<virtual>I have never gone over my transfer allotment, but I only do personal site stuff on my Linode.
16:34<FluffyFoxeh>I was more talking about the new data centres planned
16:35<Peng_>I almost went over once :D
16:35<virtual>ahhh, yeah. But they gotta compete against others who are already there. Also, if they can get ahead of the curve, even betterer.
16:35<virtual>I am glad they're in SYD, but now I have to find budget for another Linode. Budget is super duper tight at the moment, which is frustrating. :P
16:35<linbot>New news from community: How can I download a directory with many files from my Linode? <https://www.linode.com/community/questions/19270>
16:37<FluffyFoxeh>just get more money
16:37<virtual>FluffyFoxeh: working on it ;)
16:38-!-|GIG [~MYOB@158.115.253.31] has quit [Remote host closed the connection]
16:38-!-|GIG [~MYOB@158.115.253.31] has joined #linode
16:38-!-|GIG is "J" on #moocows #linode
16:48<DrJ>Not that I use excess transfer (Linode is gracious to give me way too much allowance)... but nice to see the lowered price
16:48<DrJ>you have got to be kidding me though
16:48<DrJ>"Instantaneous DNS Manager updates"
16:49<DrJ>I literally just finished moving all my DNS to cloudflare for this exact reason
16:49<DrJ>by literally I mean last week
16:50<DrJ>I've begged for "instantaneous" dns updates in here for years... finally bite the bullet and move... and not a week later linode announces they are going to roll it out
16:51<DrJ>*facepalm*
16:51<Peng_>Well, they might be aiming to roll it out in December 2020. :P
16:51<Peng_>and unless they turn off caching, it's still gonna be a *little* slower than Cloudflare.
16:58<millisa>mumbai, sydney, and toronto all get pretty patch type badges for their datacenters. I demand similar graphics for other datacenters so I can put it on my linode merit badge vest
16:58<millisa>based on the name . . .they are stickers? https://www.linode.com/wp-content/uploads/2019/12/stickers-collection-e1576718069299-1536x495.png
16:59<Peng_>You can but stickers for Atlanta from https://www.redbubble.com/shop/dumpster+fire+stickers
16:59<Peng_>buy*
16:59<virtual>LOL
16:59<millisa>!point Peng
16:59<linbot>millisa: Point given to peng. (46) (Biggest fan: millisa, total: 11)
16:59<virtual>I think I need to buy some of those and hand them out in the office ;)
16:59<millisa>(don't see the stickers on the real swag shop)
17:02<millisa>"Per customer network VLAN" that one is the one I'm most excited about
17:05<grawity>"that inter-linode networking stuff sounds awesome", I say to myself, while having only one linode total and already planning on getting rid of it as well
17:06<millisa>Kinda hoping it allows multiple vlans per account so I dont have to create more accounts to keep customers of mine isolated from each other.
17:07<Peng_>And multiple accounts per VLAN? :D
17:08<millisa>not sure I'd use that . .but I'm sure there's someone that would
17:40<nbm>i don't remember who sent me a link explaining how to install void linux with linode
17:40<nbm>i lost the url
17:41-!-koenig [~koenig@00029c1b.user.oftc.net] has quit [Quit: WeeChat 2.5]
17:42<millisa>how long ago?
17:42<virtual>whoah, this might be old news to some of you - but I just saw a bonus credit in my account!
17:42<virtual>thank you, Linode!
17:44<nbm>millisa: not so long ago
17:45<nbm>maybe 4 months ago?
17:45<millisa>if it was back in october, i vaguely remember doing one
17:45<nbm>it was probably yours then
17:45<millisa>it was mostly based on this https://wiki.voidlinux.org/Install_on_a_VPS and the linode custom distribution guide
17:47<millisa>these were my notes from then: https://vomitb.in/htbxMQyJHT
17:47<nbm>yes i think it was yours
17:48<nbm>because of the custom grub lines :)
17:48<millisa>those were mostly to get lish working
17:49<nbm>thank you millisa
19:06-!-Dataforce` [~dataforce@dataforce.org.uk] has joined #linode
19:06-!-Dataforce` is "Shane "Dataforce" Mc Cormack" on #linode #bitlbee #oftc #DMDirc
19:06-!-Dataforce is now known as Guest12744
19:06-!-Dataforce` is now known as dataforce
19:06-!-Cajs [Cajs@185.198.189.47] has quit [Read error: Connection reset by peer]
19:07-!-Guest12744 [~dataforce@dataforce.org.uk] has quit [Remote host closed the connection]
19:09-!-phyber_ [phyber@2a02:8011:25:c0de::c0:ffee] has joined #linode
19:09-!-phyber_ is "People said I was dumb, but I proved them!" on #linode
19:09-!-Cajs [Cajs@185.198.189.47] has joined #linode
19:09-!-Cajs is "Cajs" on #linode
19:09-!-|avril [wha@cute.b.oy.hu] has joined #linode
19:09-!-|avril is "..." on #tor-bots #moocows #linode
19:11-!-avril [wha@00028d51.user.oftc.net] has quit [Ping timeout: 480 seconds]
19:11-!-SpydarOO7 [spydar007@spydar007.user.oftc.net] has quit [Read error: Connection reset by peer]
19:13-!-rweir- [foobar@quandry.ertius.org] has joined #linode
19:13-!-rweir- is "no" on #linode
19:13-!-Kuukunen- [~qq@kuukunen.net] has joined #linode
19:13-!-Kuukunen- is "qq" on #linode
19:13-!-Hazelesque_ [~hazel@phobos.hazelesque.uk] has joined #linode
19:13-!-Hazelesque_ is "Hazel" on #linode #ceph #nlug #bongo
19:15-!-Spydar007 [spydar007@spydar007.com] has joined #linode
19:15-!-Spydar007 is "Spydar007" on @#Mikaela @#kovri #virt #supybot #spi #redditprivacy #perl #oftc #moocows #mastodon-administration #linux #linode @#kovri-dev #friendica #debian
19:15-!-tomaw_ [tom@tomaw.chair.oftc.net] has joined #linode
19:15-!-tomaw_ is "Tom Wesley <tom@tomaw.net>" on #debian-es #debian-django #debian #ceph #oftc #moocows #linux #virt #freenode #debian-ipv6 #bitlbee #linode #irssi #help
19:15-!-Roedy [Roedy@0002706e.user.oftc.net] has joined #linode
19:15-!-Roedy is "Roedy" on #openvas #linode #debian #OpenBSD
19:17-!-jhq [~jasper@shell.jhq.io] has joined #linode
19:17-!-jhq is "Jasper Backer (jhq)" on #virt #linode #debian #ovirt #debian-next
19:17-!-fifr_ [~fifr@math-pc1.mathematik.uni-kassel.de] has joined #linode
19:17-!-fifr_ is "Frank Fischer" on #linode
19:17-!-Netsplit charon.oftc.net <-> liquid.oftc.net quits: phyber, tomaw, Hazelesque, tomchen[m], Turl, lpalgarvio[m], frailty[m], fifr, DennyFuchs[m], Roedy-, (+13 more, use /NETSPLIT to show all of them)
19:17-!-phyber_ is now known as phyber
19:18-!-TonyL [~Tony@zeus.dedi.cakeforce.uk] has joined #linode
19:18-!-TonyL is "Tony L" on #oftc #linode #debian
19:18-!-Netsplit over, joins: dueyfinster
19:19-!-Netsplit over, joins: hawk
19:20-!-Tol1 [tol1@nokkala.info] has joined #linode
19:20-!-Tol1 is "Tomi Nokkala" on #linode
19:26-!-Netsplit over, joins: grawity
19:38<dwfreed>grawity: I mean, you need a grub.cfg, unless you like telling grub how to boot your Linode manually
19:41-!-Turl [~Turl@yotta.elopez.com.ar] has joined #linode
19:41-!-Turl is "Turl" on #ros #opensde #linode #kernelnewbies
19:44<dwfreed>Peng_: virtual: yes, Linode mentioned in the NextGen Network blog post that the entirety was built on Cisco Nexus hardware
19:46<Peng_>Juniper's case study kind of implies that Linode threw all their Cisco gear in the dumpster and went Juniper/Corero the moment after the Christmas DDoS.
19:47<dwfreed>https://twitter.com/linode/status/581100233058312192
19:50<dwfreed>Peng_: ^ considering the date on that tweet is 9 months before, that'd be a very expensive dumpster...
19:51<dwfreed>an ASR 9006 chassis *with nothing in it* is $7,000
19:51<dwfreed>and that's on sale!
19:52<dwfreed>the devices pictured are probably in the neighborhood of 50k a piece
19:53<dwfreed>now multiply that times 2, per datacenter
19:53<dwfreed>No, I doubt they got rid of all their cisco equipment, though those ASRs may be the only thing they kept
19:54<virtual>I thought it might be the other way around, they keep the nexuses, but not the ASRs, considering it's about DDoS, their juniper thing?
19:55<virtual>I didn't look ta the PDF, so...
19:56<dwfreed>the ASRs can handle multiple Tbps, far more capacity than Linode will have in the near future
19:56<virtual>sure, I was thinking about the ddos biz. never heard of corewhatever.
19:56<dwfreed>the issue is not the border, but the other edge
19:56<virtual>not sure if they only work with juniper routers, or something. unlikely, I'm sure.
19:57<virtual>oh? the customers...?
19:57<dwfreed>yes
19:57<dwfreed>and the hosts
19:57<virtual>never really considered that - I thought all the DDoS were inbound to Linode.
19:57<dwfreed>oh, they are
19:57<virtual>were they outbound from compromised hosts or something too?
19:57<dwfreed>some are, but those aren't nearly as much of an issue
19:58<dwfreed>the pipe coming into the datacenter is 200 Gbps+
19:58<Peng_>virtual: Looking at the PDF doesn't help much, since it's green and grey text on a white background. Or white text on a green background.
19:58<virtual>nice, made for readability then
19:58<dwfreed>but the pipe to an individual host is probably 40 Gbps
19:59<Peng_>Actually, wait, the PDF gives a date
19:59<virtual>surely you still want to block it at the border - as close to your source as possible (and then get your provider to block it too).
19:59<dwfreed>when you're getting an 80 Gbps DDoS, the border doesn't notice, but the host does
19:59<virtual>provider(s)
19:59<dwfreed>sure, which is why you just BGP the target IP on the ASR to the DoS mitigation box
19:59<Peng_>"Spataro discovered a solution at Juniper’s 2018 NXTWORK conference. The addition of Corero SmartWall Threat Defense Director, in combination with the MX960 routing platforms that Linode already used, would provide real-time DDoS mitigation."
20:00<virtual>wonder why juniper then, and not a cisco ddos device (I'm assuming one exists)
20:00<virtual>oh, they already use mx960s? wtf are the ASRs for??
20:00<Peng_>virtual: Three and a half years between the ASRs and the MX960s
20:01<dwfreed>so they may have switched to MX960s after a few years
20:01<dwfreed>I'm just on the outside these days, no idea
20:01<virtual>aren't they approximately similar devices?
20:01<dwfreed>yes
20:02<dwfreed>NXTWORK 2018 was early October 2018
20:05<virtual>I haven't had to buy any new gear since 2014. or was it 2013. now I'm just a user. :P
20:05-!-nate [NBishop@00013625.user.oftc.net] has quit [Quit: (?°?°)?? ???]
20:11-!-nate [NBishop@d-207-255-41-254.paw.cpe.atlanticbb.net] has joined #linode
20:11-!-nate is "Nathan" on #linode #php
20:12<millisa>i bought new gear today... hadnt updated my primary workstation in 4 years. this was going to be my second gen wall mount/art display type system.
20:13<millisa>had just finished mounting the motherboard and gigantic heatsink and was going to finish tightening the recessed screws. turned my back and heard a terrible crash.
20:13<virtual>no!
20:13<millisa>i guess i had just barely balanced it on the desk and the whole thing, keeled over, landing on the heatsink.
20:13<virtual>:(
20:14<millisa>it's now got a nice 15 degree tilt to it. still boots. so i've got that going for me
20:14<virtual>hmm
20:14<virtual>I guess tilted is still art.
21:04-!-Parinioa [~oftc-webi@2606-a000-1319-40d3-60bc-74ca-1a54-ef27.inf6.spectrum.com] has joined #linode
21:04-!-Parinioa is "OFTC WebIRC Client" on #linode
21:09-!-|GIG-1 [~MYOB@158.115.253.31] has joined #linode
21:09-!-|GIG-1 is "J" on #linode #moocows
21:09-!-|GIG-1 [~MYOB@158.115.253.31] has quit [Remote host closed the connection]
21:16<DrJ>ty Linode for a great decade! Been a customer since 2010... ready for another full decade :)
21:20<@jyoo>Happy to have you! :)
21:22<Parinioa>It seems there was an issue in Atlanta about an hour ago, resulting in a system reboot. Where can we get more info on this than the generic support ticket gave?
21:23<millisa>if it impacts multiple people, they usually put something up on the status page.
21:23<@jyoo>I'm taking a look at Atlanta now, please hold
21:23<millisa>you can reply to the ticket they generated and ask for more details. or give 'em a call
21:24<@jackley>if it resulted in a reboot, it sounds like a hardware issue
21:24<@jackley>which isn't something we'd normally put on the status page
21:25<@jackley>Parinioa: is your host h689-atl1?
21:25<@jyoo>Parinioa: A host in Atlanta was down for emergency maintenance about an hour ago due to a software issue. The host stopped responding and required a reboot to bring it back up.
21:25<Parinioa>The ticket just referred to "an issue affecting the physical host"
21:28<millisa>(where does the new manager show the host system's name?)
21:29<Parinioa>I was about to ask the same question. I can't seem to find it
21:29<millisa>in the old manager its on the linode details page
21:30<millisa>in that right hand sidebar, down near the bottom
21:30<@jackley>it's only in the classic manager, it doesn't exist in Cloud (I asked that b/c I thought you might be using classic)
21:31<Parinioa>I'm happy to use classic, but that option doesn't seem to be on the login screen anymore
21:31<millisa>manager.linode.com
21:31<@jackley>ya – but we only had an issue with a single host in ATL in the last hour
21:32<@jackley>so I guess there's no need to confirm
21:32<Parinioa>It is currently on h689-atl1 (KVM)
21:32<@jyoo>Yep, that one had an RCU stall that caused it to go down briefly
21:33<Parinioa>That helps, thanks.
21:35<Parinioa>Though it makes me worry that loosing the classic manager also means we're loosing some functionality. Are their plans to impliment all the info in the classic manager into the cloud manager eventually?
21:37<@jackley>ya we're aiming to have parity for most things in the cloud manager (we just added longview, which was one of the last big features we needed)
21:39<@jackley>I don't believe we're planning on adding host info to the API, though, so that will likely not be there by the time the classic manager is EOL'd
21:39<Parinioa>That will make answering your earlier question quite a bit harder.
21:41<@jackley>true! but I also could have provided an answer without having to ask that
21:42<Parinioa>No doubt. I"m just a grumpy old man that doesn't like to see things change when they were working for me as is.
21:42<@jackley>heh I can empathize with that
21:45<LouWestin>I was just reading Linode’s blog recap for the year. I didn’t know Linode had added DDOS protection. Least not general blanket protection
21:47<virtual>I think it's probably in Linode's best interests to do this too - those DDoS's impacted more than just the targetted customers of Linode - if they were targeting those and not Linode infrastructure.
21:47<virtual>Depressing that it's even a thing that has to exist though.
21:48<LouWestin>I believe it was about 2 years ago when Linode got hit really hard for two weeks . Atlanta was probably hit the worst
21:49<millisa>4 years ago
21:49<millisa>xmas of 2015
21:49<millisa>https://www.linode.com/2016/01/29/christmas-ddos-retrospective/
21:49<Peng_>Time flies when you're not down :D
21:50<LouWestin>Essh, indeed!
21:51<LouWestin>That means I’ve been a linode customer for...5 years at least
21:52<LouWestin>Probably 6. Too lazy to login to my account right now
21:53<@jackley>feels like it was so long ago
21:55<LouWestin>I’m a bad judge of age. lol hence being off by two years.
22:24-!-koenig [~koenig@00029c1b.user.oftc.net] has joined #linode
22:24-!-koenig is "koenig" on #linode
23:02<Nightmare>I remember that blog post after Linode HQ got swatted, lol
---Logclosed Tue Dec 31 00:00:48 2019