|
larchesdanrew posted:She did not last the day.
|
# ? Aug 22, 2019 22:09 |
|
|
# ? May 4, 2024 11:13 |
|
larchesdanrew posted:I feel pity and relief.
|
# ? Aug 22, 2019 22:16 |
|
larchesdanrew posted:I can't be arsed to go back and dig up the old post, but a few years ago I was involved in some drama involving the wicked Property Control Clerk, a staggeringly poorly documented equipment disposal, and the most lies I've ever seen a human being tell. I almost lost my job and got into serious legal trouble over the property clerk throwing mountains of buses on top of me to protect her own rear end. I remember the previous interaction, and I feel triumph. but it isn't mine so I'm letting it go. Glad that one doom is off your shoulders.
|
# ? Aug 22, 2019 23:04 |
|
Allow yourself a few fist pumps, and some private mockery. You've earned this, friend. as an aside, "Private Mockery" would make for a good rum label
|
# ? Aug 22, 2019 23:53 |
|
larchesdanrew posted:She was not slain by a carefully laid trap from yours truly. larchesdanrew posted:He came to ask me what my experience with her was, to which I responded that it has been resoundingly negative and that I fear for my job every time inventory is involved as there are documented cases of her altering paperwork and destroying documents to protect herself.
|
# ? Aug 23, 2019 01:06 |
|
Renegret posted:That's what I was thinking. Every time I had a computer running without an hsf, it just shut down within minutes. When I've had the HSF fall off in other computers in my school, that's what happened. They'd be working slowly, then just shut off. That'd usually clue me in pretty quickly that it's a thermal problem. I'm guessing since I wasn't trying to actually do anything but wait for whatever the computer was trying to do to finish, and it was only on for less than 10 minutes, and there's a decent amount of airflow in the case, it wasn't quite enough to reach the shutdown threshold. Still, running at 3.2ghz processor at 7% its rated speed... Never experienced a Pentium II 266 running Windows 10! I decided I couldn't be bothered cleaning Chromebooks today and instead helped put together a lovely filing cabinet and disassembled a large laminator to unjam it. Did deliver all those iPads, though. Re: Inventory stuff. Our Accounting person is responsible for inventories, and she's pretty paranoid about it. Spent weeks trying to find three Toshiba docking stations, then since they were under the $200 threshold, she relented and just removed them all from our inventory. Our original inventory was hackneyed when I started there, and lots of stuff wasn't inventoried at all, and what was, was done strangely. I took it upon myself to start my own excel inventory, and a year later the district finally decided to get asset tracking software. We're about to move to our third inventory system in three years. Samanage (it literally duplicated our entire inventory last year somehow), then Follett Destiny (it's a great library checkout system, but the tech inventory is clearly tacked on), and now we're moving to Asset Panda. Pyroclastic fucked around with this message at 02:23 on Aug 23, 2019 |
# ? Aug 23, 2019 02:16 |
|
A few hundred tickets came in two days ago...to the lobby of the theater our worst client uses for their shows because they're a theater nonprofit. And then those tickets got torn because the ticket scanners (which are jumped up Android phones with a barcode scanner glued on and cost $1k each) wouldn't connect to the WiFi we just upgraded so they couldn't scan tickets. More specifically, we just put in Meraki MR45s to replace the MR42s that were there, and it does seem like the 45s are much pickier about what devices they allow to connect (or I guess the devices have a problem negotiating with them). From a quick Google I see some references on Reddit to old Intel chipsets having problems but not a lot else. I eventually got the scanners to work yesterday by setting the SSID to "full compatibility" in the Meraki controls, which amusingly states "compatibility back to pre-1999 devices", but we're also seeing issues with their laptops because of course they have incredibly ancient machines trying to connect. Anyone else seen connection problems with MR45s, and/or have any suggestions beyond upgrading drivers? I'd like to think Meraki will release a firmware update at some point to help fix these issues but my experience with them is that firmware updates are not released quickly. In the meantime apparently the scanners aren't connecting at the same distances they used to (possibly switched to 5 GHz for some reason, and the theater is all concrete so that's fun) so I guess I'm heading over to the theater today again - normally one of our techs would but I live very close to the place - to gently caress around with Android 7 phones, thanks Zebra.
|
# ? Aug 23, 2019 19:52 |
|
Aerohive used to have very specific SSID profiles if you needed old handheld inventory devices to work - stuff like turning off any 'advanced' features like power saving, letting 802.11b devices connect etc.
|
# ? Aug 23, 2019 20:09 |
|
Our older Meraki stuff seems to work but anything purchased post-2016 seems to have issues, either sporadically or just all the time. None of our brand new HP PCs and no iPhones can connect to any of our SSIDs in one of our new facilities, but my Dell computer and Android phone connect just fine. I'm far enough removed from dealing with that stuff that I don't care any more but yeah for anything where you have to deliver stable wireless, Meraki is not going to fit the bill unless you have total control over the devices connecting and can do a shitload of proof of concept testing before deployment. Dealing with Meraki issues isn't a constant thing but it happens often enough that I'm bitter about the whole experience. We actually have a bunch of lovely Linksys wireless APs running in production right now because of this. Edit: AFAIK there is no fix other than to swap hardware and pray if you've already made the obvious compatibility-related changes on the dashboard and that didn't resolve it. We did have very minor improvements turning off band steering and splitting 2Ghz and 5Ghz to separate SSIDs but that's the only not-totally-obvious suggestion I can make. Sheep fucked around with this message at 20:59 on Aug 23, 2019 |
# ? Aug 23, 2019 20:55 |
|
Hrm interesting. We haven't really had issues with Meraki wireless before the MR45s, and even with the 45s it seems like it's old stuff having issues, not newer (<2 years old) equipment. I ended up fully "updating" the ticket scanners to their most recent patch level, which was July 2019 except they're still running Android 7 But whatever, that seemed to help the disconnection issues. Now of course there's the problem that we need another AP in the building because there's a solid 2 feet of concrete between the first and second floors (with giant steel beams embedded in the concrete). But of course the same construction also makes it difficult to run new wires. Anyway it's now 100% beer o clock so yeah no fucks given until Monday.
|
# ? Aug 24, 2019 02:18 |
|
A ticket did not come in, instead a giant pile of printouts appeared on my desk. Changes for one of the web pages, one that notoriously has had errors when it's been updated.
|
# ? Aug 24, 2019 02:24 |
|
It really sounds like you need a proper wireless survey and then to design your network around the results of that and the radio specifications of the terminals you're connecting to it.
|
# ? Aug 24, 2019 16:49 |
|
Maigius posted:A ticket did not come in, instead a giant pile of printouts appeared on my desk. Changes for one of the web pages, one that notoriously has had errors when it's been updated. You get printouts? Not a photocopy of scrawlings made on a legal pad?
|
# ? Aug 24, 2019 18:05 |
|
Thanks Ants posted:It really sounds like you need a proper wireless survey and then to design your network around the results of that and the radio specifications of the terminals you're connecting to it. Yes, I know, and in fact will be conducting such a survey real soon now. The annoyance here is that there was an existing MR42 AP that we replaced (same exact spot), and that worked fine with everything including these ridiculous scanner devices, and I was inquiring if anyone had had specific experience with the relatively new MR45s as to whether they were worse than the 42s. Specifically at listening, since the transmit signal level is low but not non-existent and my supposition is that the extremely weak response from the scanners was picked up well enough by the 42 but isn't by the 45. But frankly a wireless survey won't really help much in that it will almost certainly just confirm my suspicions of just how low the signal level is (I mean, I can get an idea using my phone's signal analyzer, the location with problems is around -78 dB) and will tell me I need to put an AP in on the second floor where the return signal won't be blocked by two feet of concrete and steel beams. Which, I already know. And which I can't do jack poo poo about because it's not feasible from a monetary standpoint to run the wires because of the lack of access through the building (and I really would rather not mesh APs).
|
# ? Aug 24, 2019 23:52 |
|
I have 1500 alert emails. A guy on my team was doing some reboots this weekend. Excited to see if it went well!
|
# ? Aug 25, 2019 22:51 |
|
If you'd like to read the post-mortem from my company's ransomware incident, shoot me a PM and I'll send the link your way. I don't think there are any problems with me posting it right here, but I'm playing it safe!
|
# ? Aug 26, 2019 13:26 |
|
If the postmortem is publicly posted, the only problem I can imagine with sharing the link is we'll know where you work. Which, fair.
|
# ? Aug 26, 2019 15:43 |
|
Zero VGS posted:Seems to be something having to do with OneDrive; deleting a folder in the O365 web portal will cause OneDrive to delete the folder on my end. The few folders I have that aren't synced to OneDrive can be deleted normally. At least now I can get rid of the huge iCloud backup I took for a user. I had this happen on a fresh Win10 install when my time zone synchronized and put the system clock way ahead of where it was on initial install. I'd whack delete on local folders and nothing would happen as the one drive integration freaked out over timestamps from the future. Did anything happen to the system time by chance?
|
# ? Aug 26, 2019 16:53 |
|
NOC calls me at 3am: Hey, <teamA> needs a few servers restarted, can you assist with that? Me: Sure, that's what I'm here for. Join me to the Teams room Me in Teams room: Ok, so who are we hitting with the hammer tonight? Team A Person: So, we're getting a lot of connections from <Environment> Me: ... Team A Person: ... Me: And we believe rebooting some of the <TeamA> servers will mitigate/resolve this? Team A Person: Ok Me: That was a question. Team A Person: Oh, yeah. That's what we've done in the past to resolve this Me: Ok, which servers are we looking to reboot Team A Person: All of the prod hosts for <TeamB>'s environment Me: So... yeah. I'm not gonna go rebooting <TeamB>'s servers at 4am at the behest of another team. Let's think about engaging that team first, eh? But yeah, chief, I'm gonna just go hitting someone else's servers with a hammer. Not to mention, it's 99.999% more likely that a reboot isn't necessary, just cycling a service or 5.
|
# ? Aug 27, 2019 09:12 |
Rebooting demonstrably doesn't fix anything, it only masks the problem. System-level tracing with minimal performance hit to do in-place troubleshooting, or even just capturing the whole operation to do post-mortem analysis. DTrace is capable of doing this, and on top of only taking ~2% of cputime it only does it for the code that's being actively traced - and best of all, it's available on all platforms: Windows, Linux, macOS, FreeBSD and NetBSD, Illumos, Solaris.
|
|
# ? Aug 27, 2019 09:29 |
|
D. Ebdrup posted:Rebooting demonstrably doesn't fix anything, it only masks the problem. Sure, but I leave that to the app developers to handle after the fact. I'm not in the business of debugging someone else's code, especially at 4:30am now. And yeah. Rebooting, in the *nix world in particular, is a last resort to get poo poo moving if all else fails.
|
# ? Aug 27, 2019 09:32 |
|
AlexDeGruven posted:Sure, but I leave that to the app developers to handle after the fact. I'm not in the business of debugging someone else's code, especially at 4:30am now. reboot a N*X server means a hardware or kernel issue. or lovely java is leaking uncontrollably again
|
# ? Aug 27, 2019 11:10 |
RFC2324 posted:reboot a N*X server means a hardware or kernel issue.
|
|
# ? Aug 27, 2019 11:44 |
|
D. Ebdrup posted:Java belongs in a jail, with resource control via rctl. I don't get to engineer these lovely platforms, I just get to keep them alive
|
# ? Aug 27, 2019 11:48 |
|
RFC2324 posted:reboot a N*X server means a hardware or kernel issue. Java is never not leaking or lovely. RFC2324 posted:I don't get to engineer these lovely platforms, I just get to keep them alive Samesies
|
# ? Aug 27, 2019 11:48 |
RFC2324 posted:I don't get to engineer these lovely platforms, I just get to keep them alive
|
|
# ? Aug 27, 2019 11:57 |
|
AlexDeGruven posted:Java is never not leaking or lovely. Tell me about it. I play a lot of Minecraft in my off-time, and because I use mods, I'm stuck on the Java version. Every week or so I have to reboot my PC because over half my memory is "in use" due to Java not releasing it properly.
|
# ? Aug 27, 2019 14:13 |
|
D. Ebdrup posted:Rebooting demonstrably doesn't fix anything, it only masks the problem. The moment it goes from "system hiccup" to "the problem happened again" looking into the cause should be mandatory. If it happens a second time, it's going to happen a third.
|
# ? Aug 27, 2019 15:49 |
klosterdev posted:The moment it goes from "system hiccup" to "the problem happened again" looking into the cause should be mandatory. If it happens a second time, it's going to happen a third.
|
|
# ? Aug 27, 2019 17:19 |
|
Co-worker who recently transitioned to a different team: Did we ever post-mortem the most recent spaghetti incident? Me: Post-mortem? POST-MORTEM? Do we EVER post-mortem anything? Co-worker: So that's a no. Me: I mean, I asked for one, but you know no one in HQ can hear voices that are higher pitched than their own. Meanwhile, ridiculously short handed because of recent staff transitions and several people burning out and quitting, and I've got a bunch of net new sites crawling up my rear end about getting imaging online. This is fairly trivial on my end, in terms of effort, but the problem is trying to unfuck an automation script written by a former teammate who had never written code before while also getting the prod environments online. (He wrote a function...wherein all the variables get declared. Send help, I'm officially dead.) Again, trivial effort, but time consuming, so I need everyone to just get out of my loving way and let me do it. Instead, the most senior member of the senior team says, "We can dogpile this and get it done faster!" Like... My dude. I just told you that's the opposite of helpful. Plz stop.
|
# ? Aug 27, 2019 17:29 |
|
Wheeeeee. It's busy season again and our dining operation's credit card processor has been up and down since yesterday. Can't even access their status page and their support line has been busy for an hour. Thank god kitchen people will just nod when you tell them "<processor>'s poo poo is hosed" and not expect me to fix a major company in another state's systems with magic like some other departments.
|
# ? Aug 27, 2019 22:41 |
|
monsterzero posted:Wheeeeee. It's busy season again and our dining operation's credit card processor has been up and down since yesterday. Can't even access their status page and their support line has been busy for an hour. Thank god kitchen people will just nod when you tell them "<processor>'s poo poo is hosed" and not expect me to fix a major company in another state's systems with magic like some other departments. I'm sure this is why I had to go to three different liquor stores to buy whiskey last night. Credit card machine down in both the first two.
|
# ? Aug 28, 2019 19:38 |
|
RFC2324 posted:reboot a N*X server means a hardware or kernel issue. Or the find command forks itself to eternity and fills up 256gb of memory and 512gb of swap.
|
# ? Aug 28, 2019 20:34 |
|
monsterzero posted:Wheeeeee. It's busy season again and our dining operation's credit card processor has been up and down since yesterday. Can't even access their status page and their support line has been busy for an hour. Thank god kitchen people will just nod when you tell them "<processor>'s poo poo is hosed" and not expect me to fix a major company in another state's systems with magic like some other departments. I work for a chain of retail stores.. Its always fun when Verifone's gateway service goes down (Verifone Payware Connect) Last time it happened it was down for nearly an hour, in the middle of the day. And it never fails there are always several transactions in progress when this happens. We end up with the problem where our POS software submitted a transaction, but never received a response back. So we have no idea if a customer's card was actually charged or not. Sometimes we find they where double charged. Always fun cleaning those up the next day.
|
# ? Aug 28, 2019 21:20 |
|
mood: https://twitter.com/mrfixit42/status/1166874014540677120
|
# ? Aug 29, 2019 01:45 |
|
D. Ebdrup posted:Java belongs in a jail, with resource control via
|
# ? Aug 29, 2019 01:56 |
porktree posted:Or the find command forks itself to eternity and fills up 256gb of memory and 512gb of swap. If you're doing a proper job of it, FreeBSD need only be rebooted for kernel upgrades. Although I do wish someone would finish the kload project (like ksplice for RHEL/Oracle Linux, as FreeBSD has stable ABI/KBI by design), since that'd make it so that you would only need to reboot every 5 years when a major supported branch goes EOL. mllaneza posted:A lack of about killing Java I don't think I've ever felt so good about decommissioning a server and bringing it up again for the new setup, but it was so cathartic to finally get rid of it after fighting with it for over 3 years.
|
|
# ? Aug 29, 2019 11:32 |
|
porktree posted:Or the find command forks itself to eternity and fills up 256gb of memory and 512gb of swap. As an AIX person, the idea of that much swap gives me the shivers. I run 16gb on my 256gb boxes. If AIX is paging, something is wrong.
|
# ? Aug 29, 2019 15:01 |
|
Our hosting provider of 6ish years is closing a datacenter this Sunday and I've spent the last 9 months migrating what is probably 1/3rd of our customer base to Azure or another datacenter with the same provider. There is only one fucker left and they are insisting we let them continue using it in-place until noon Saturday. I want to die suuma fucked around with this message at 02:55 on Aug 30, 2019 |
# ? Aug 30, 2019 02:42 |
|
|
# ? May 4, 2024 11:13 |
|
suuma posted:Our hosting provider of 6ish years is closing a datacenter this Sunday and I've spent the last 9 months migrating what is probably 1/3rd of our customer base to Azure or another datacenter with the same provider. Oh no they shut it off at Noon on Friday, I migrated you by 4 PM you're welcome!
|
# ? Aug 30, 2019 02:54 |