|
Loten posted:UPS job OK turns out this guy is loving nuts. Had more dramas with him today after the CIO straight up forwarded my email about the incident to the user. He then reported a very distorted version of events, which I won't bother going into as it's not that exciting, but I'm sure there is more to come because after talking with with my manager today the guy has previously: - Asked for a work phone. Been given the only type of phone we have (and the only type approved by finance), which happens to be an older Nokia phone. He then proceeded to stomp on it and stab it with a pen. Wrapped it up in bubble wrap and sent it to the support desk with a note along the lines of "Don't send me this crap". - There used to be a poster with all the faces of the IT staff hanging in the IT area. If he wasn't satisfied with your performance/response to his issues, next time he visited head office he would come over and draw a big cross over that staff members face. - Apparently has "a list" of people he doesn't like in his desk.
|
#
?
Jun 17, 2014 11:45
|
|
|
|
| # ? May 21, 2024 15:59 |
|
|
Wow, what a tremendous dickhead!
|
|
#
?
Jun 17, 2014 11:47
|
|
|
Write his name on the list!
|
|
#
?
Jun 17, 2014 11:55
|
|
|
Loten posted:- Apparently has "a list" of people he doesn't like in his desk. I think his "list" just got a little longer starting today.
|
|
#
?
Jun 17, 2014 11:55
|
|
|
Loten posted:- There used to be a poster with all the faces of the IT staff hanging in the IT area. If he wasn't satisfied with your performance/response to his issues, next time he visited head office he would come over and draw a big cross over that staff members face. Wow. A hypothetical reasonable person might very well interpret that as a threat, considering his demonstrated aggressive tendencies. Said hypothetical reasonable person might then take that up with HR... ...and probably suffer harassment both from the jerk and management over it, because everyone knows "rockstars" need to be pampered and given every indulgence to continue to create value for the company. 
|
|
#
?
Jun 17, 2014 12:25
|
|
|
I'd bet he's actually the worst in his company as far as performance but manages to bitch out his own boss until the guy gives him a positive performance review.
|
|
#
?
Jun 17, 2014 12:48
|
|
|
He manages the one department/sub section of the business that is profitable at the moment. I'm fairly sure he's untouchable but it will be interesting to see how this plays out.
|
|
#
?
Jun 17, 2014 12:53
|
|
|
Do you have any sacrificial equipment you could give him? Maybe an old decommissioned printer that you can send him and tell him to smash that instead? It could be a win win situation where you just send stuff to him for disposal, and he gets to satisfy his thirst for electronic destruction.
|
|
#
?
Jun 17, 2014 14:20
|
|
|
A laptop came in. It won't power on and the user has stuff on the HDD they need. It's a Vostro 3450. OK, let's get that drive out...  gently caress you, I'm a Vostro!   Even if you tried taking pliers to the plastic bits covering the drive, you wouldn't be able to get to that top-left screw or have enough wiggle room to pull it from the connector. We only had 3 of these in the whole company and one of its brothers did the same thing. Bonus: those sticky notes on the palm rest are all the user's usernames and passwords
|
|
#
?
Jun 17, 2014 15:40
|
|
|
Possible  !I have been feeling super discontent and overwhelmed at my current job lately. Tons of projects with poor management, scope going out of control, and 'not enough budget' to give me a promotion after 5 years, but plenty to hire a 4th Senior VP. So I started throwing resumes into the wind, not really expecting anything to happen. In the past 2 weeks I've finally started getting callbacks - 1 from another financial institution and another internal opportunity doing IS Risk. I went for an interview yesterday and it went really well. I'd be able to learn some SQL and *nix stuff, and work downtown so I get to take the train. And my interview for the risk position is tomorrow. I am already on good terms with the manager of that department, and I would be doing all sorts of fun stuff like managing GPOs and AD, certificates and PKI, and doing a proper WSUS setup, with the added bonus of a 12% raise. Now I'm conflicted. When it rains it pours I guess.
|
|
#
?
Jun 17, 2014 15:59
|
|
|
bitterandtwisted posted:gently caress you, I'm a Vostro!  Here's a 3460
Roargasm fucked around with this message at 16:04 on Jun 17, 2014 |
|
#
?
Jun 17, 2014 16:00
|
|
|
bitterandtwisted posted:Bonus: those sticky notes on the palm rest are all the user's usernames and passwords
|
|
#
?
Jun 17, 2014 16:20
|
|
|
A ticket came in a few days ago with the text simply reading, "Why did you remove my photos?". Turns out the user got hit with cryptowall almost a month ago and just now noticed that her files weren't loading. Never got the extortion screen, and the user is a tiny old lady who basically wouldn't notice if the thing were on fire. And, of course, Mcafee didn't catch a thing either because why the hell would it. I mean, it's not like we need proper security or anything. We're just an airport located next to a major metropolitan center, no big deal. The Muffinlord fucked around with this message at 17:09 on Jun 17, 2014 |
|
#
?
Jun 17, 2014 16:37
|
|
|
The Muffinlord posted:A ticket came in a few days ago with the text simply reading, "Why did you remove my photos?". McAfee, Endpoint, etc are all just about useless at this point. I really wonder if we would even use Endpoint if it didnt come free. I think at some point it will just make more sense to have good group policies, user training, and a great IPS/IDS system setup.
|
|
#
?
Jun 17, 2014 17:10
|
|
|
It'd be pretty cool if our engineers and customers could communicate a little better on what sort of computer equipment will be required for the tests we run for them and just how the tests are supposed to function. It's getting old have to babysit engineers who suddenly require a brand new machine, printer and the works on short notice and then need help setting everything up because the customer provided zero supplies or support. And then troubleshooting the problems with the test because either the test instructions are wrong or they've provided a bunch of hardware that doesn't work properly together. Sometimes I wonder how anything gets done. (It doesn't.) I'm also not sure whether I should truly be annoyed with the engineers or the idiot sales guys who set up these contracts. (Probably both.)
|
|
#
?
Jun 17, 2014 17:50
|
|
|
A ticket came in, customer is upgrading MSSQL server from express to standard. Fair enough, express has plenty of limitations and standard will give them lots of room for growth. Turns out the company they purchased their licenses from sold them on the idea of making it a failover cluster as well in case something goes wrong. Again, fair enough, can't be too careful I guess, never had setup one before so do some digging around to find out what to do, seems to be straightforward. Then I notice on the MSDN articles that a clustered MSSQL enviroment requires Server Enterprise or Data Center editions. They were sold Server 2008 R2 Standard. 
|
|
#
?
Jun 17, 2014 18:58
|
|
|
Alighieri posted:A ticket came in, customer is upgrading MSSQL server from express to standard. Fair enough, express has plenty of limitations and standard will give them lots of room for growth. Turns out the company they purchased their licenses from sold them on the idea of making it a failover cluster as well in case something goes wrong. Again, fair enough, can't be too careful I guess, never had setup one before so do some digging around to find out what to do, seems to be straightforward. Then I notice on the MSDN articles that a clustered MSSQL enviroment requires Server Enterprise or Data Center editions. They were sold Server 2008 R2 Standard. How did they even get a 2008 R2 license? I'm going to assume either A. It was volume license with downgrade rights or B. They are shady as gently caress.
|
|
#
?
Jun 17, 2014 19:00
|
|
|
m.hache posted:How did they even get a 2008 R2 license? I'm going to assume either A. It was volume license with downgrade rights or B. They are shady as gently caress. The licenses didn't come from our company so I have no clue how they got a hold of them. Went ahead with just separate SQL servers and have one replicating the other. It's not a failover setup so if the main server goes down settings will need to be changed, but better then a lot of clients that have no backup plan at all and just say #yolo.
|
|
#
?
Jun 17, 2014 19:11
|
|
|
#yolo is the standard approach for small businesses. Business essential files without backups? CHECK Unique snowflake database without backups? CHECK On failing old poo poo computer used only for that one purpose? CHECK Refuses to spend money? CHECK Thankfully I am not responsible for that junk.
|
|
#
?
Jun 17, 2014 19:18
|
|
|
I know everything is moving to the cloud and every drat site I go to people are all "You should move your poo poo to the cloud yo. Take away the management headaches". And then here I am, our online tracking system has been flaky as hell for 5 days with no repair in site leaving us high and dry. My users are on my case every hour on when it's going to be fixed and I just have to sit here with my thumb up my rear end while I wait for them to fix it.
|
|
#
?
Jun 17, 2014 19:34
|
|
|
You can do mirroring or log shipping replication on 2008 R2 Standard btw, it's not clustering or anything but it's better than nothing. Hell, the witness for the mirror can be an express instance so there's no additional licenses required.
|
|
#
?
Jun 17, 2014 19:42
|
|
|
m.hache posted:I know everything is moving to the cloud and every drat site I go to people are all "You should move your poo poo to the cloud yo. Take away the management headaches". This has less to do with 'the cloud' as a concept and more to do with whoever your provider is. You could just as easily have an on-site service falling over whilst their support buries their thumbs in their collective asses.
|
|
#
?
Jun 17, 2014 19:49
|
|
|
Caged posted:This has less to do with 'the cloud' as a concept and more to do with whoever your provider is. You could just as easily have an on-site service falling over whilst their support buries their thumbs in their collective asses. This is true. It's just when people ask why I decided to host my email internally it's examples like this that I like to bring up. If it breaks at least I can work on something to fix it. I guess what I'm saying is I hate sitting around when there is a problem and being unable to work on it.
|
|
#
?
Jun 17, 2014 19:55
|
|
|
Sometimes it's not a bad thing to push the blame to the external provider "Yes email is down, it's on Microsofts server so yes we can't do anything about it".
|
|
#
?
Jun 17, 2014 19:57
|
|
|
GreenNight posted:Sometimes it's not a bad thing to push the blame to the external provider "Yes email is down, it's on Microsofts server so yes we can't do anything about it". The problem with this is that then they get mad at you for choosing 365 as a service provider even if it was their idea in the first place.
|
|
#
?
Jun 17, 2014 19:58
|
|
|
In my experience, clolud is fine and dandy when it works, when stuff starts doing weird poo poo is when the cloud is a goddamned nightmare. I literally spent a month on the phone with Microsoft support when quite a few of my internal users were becoming randomly unlicensed when I had to support o365. It took them that literal month to admit that the cluster was hosed and made me sit with my thumb up my rear end going through the motions.
|
|
#
?
Jun 17, 2014 20:00
|
|
|
Riso posted:#yolo is the standard approach for small businesses. I've had to deal with this for three years in my previous job as a sysadmin in the SME segment. I took care of a bunch of different customers, and this happened at almost all of them. Nothing quite like sitting around in the evening waiting for a slowly dying server to finish CHKDSK so it could continue booting properly for another few weeks. Also another company had a failing NT4 shitbox that hadn't been rebooted in four years just to keep track of clockins. Thankfully it was replaced before it needed to be moved to a new location because it wouldn't come up afterwards.
|
|
#
?
Jun 17, 2014 20:10
|
|
|
Riso posted:#yolo is the standard approach for small businesses. For as funny as that is, I'm loving terrified of ever being in a situation to support these. You know when they do poo poo the bed, you know who's going to get blamed for it. Renegret fucked around with this message at 20:39 on Jun 17, 2014 |
|
#
?
Jun 17, 2014 20:25
|
|
|
Bonus: I learned the same company once had, because of lack of back ups, to hire a data recovery company for lots of $$$ to try and restore critical files from a dead HD. Surprise, the files were unrecoverable and they still haven't learned a thing!
|
|
#
?
Jun 17, 2014 20:35
|
|
|
An email quarantine notification came in. Some manager accepted an interview at another company via her work email, and mistyped her phone number in the form of a social security number so it got flagged. I had to read bits of the email to confirm it wasn't violating policy somehow before I sent it out. The idiot even started the entire email conversation chain with her personal email, then told the HR person to start responding to her work address instead since "I check that one more often". The biggest insult is she (claims she) makes almost as much as I do. 
|
|
#
?
Jun 17, 2014 20:51
|
|
|
...a while ago. (not in my supervisory chain) Remember all those unapproved/obsolete forms that people at one location keep asking for? Here they are!  Uh, OK. Hey, (direct supervisor), is all this legit to use? I know <specific forms> are obsolete and will attract auditor spankings. (direct supervisor) Well, I guess so, right? Maybe you could check with (department head)? 'cause it would suck to 1) get dinged and 2) have to redo everything with the correct versions. Sure, I'll forward these on and ask!Today:  (department head) Here's these same forms that were forwarded on to me, we can use them, toodles!  Email with 30 attachments, originally packaged in August of last year, using non-current versions even at that time. About ten currently obsolete versions - three of which are audit-checks/loving legally required to have been updated back in November, one with a hand-written notation that it's a draft, one that's a WIP that was rolled back to a different design entirely and isn't even really a draft yet anyway, *and* is the responsability of a different department entirely... How long do I need to hold onto the fuser (Toshiba eStudio 6550c) to ensure being out of the office when bus-undercarriage inspection time rolls around? To: CC:  (different department head consulted re: audit-check forms)Subject: poo poo be hosed yo <detailed description of the flaws, with attachments> "Please confirm that I am to use this particular formset." Respec... (are you loving kidding me?) (Can't wait for 'em to pull this kind of poo poo with the actual legal documents/court paperwork/poo poo that lawyers will gaze upon with intellects vast, cool and Hmm, I hear there's a job opening up in Edmonton.
|
|
#
?
Jun 17, 2014 20:56
|
|
|
Sirotan posted:An email quarantine notification came in. Some manager accepted an interview at another company via her work email, and mistyped her phone number in the form of a social security number so it got flagged. I had to read bits of the email to confirm it wasn't violating policy somehow before I sent it out. The idiot even started the entire email conversation chain with her personal email, then told the HR person to start responding to her work address instead since "I check that one more often". We had one legendary fuckup here who once pasted his entire cover letter into a ticket's work log. He quietly closed that ticket and it managed to slip by management without anyone important noticing. However, it's a legend that's been passed down through the generations here, even I've seen the ticket even though he left for unrelated reasons years before I was ever hired. After all these years, it's still funny.
|
|
#
?
Jun 17, 2014 21:22
|
|
|
A sev 1A-ticket came in, complaining about crippling slowness and freezing in a literally life-and-death application. How to tell you might have a CPU bottleneck in your database, this is a Profiler trace for queries using more than 100ms of CPU-time.  During peak:  After peak:  Same query, same database, 45 minutes difference. Duration is microseconds, CPU is milliseconds. In two seconds it recieved 8 batches, each requiring 3 seconds of CPU time. The server had 4 CPUs. You do the maths. During peak this thing waited for CPU time for 19 seconds.  Oh and the (old-rear end) application does data retrieval synchronously, so that means the GUI will hang untill the query completes. Solution (i.e. rearranging the deck chairs on the Titanic): Migrate other VMs off the host untill there's one-to-one in vCPUs to physical CPUs ensuring this special snowflake database always has available physicals, oh and allocate another vCPU to the VM as well bringing it to 5 and the user experience from "unusable" to "lagspikes". At least untill peak time tomorrow. I can't wait. God help us if this thing ever fails over to another host and needs to share hardware. Time to yell at the vendor to optimize their drat SQL I think.
|
|
#
?
Jun 17, 2014 22:55
|
|
|
Caconym posted:
Ahahahahahaha, good luck.
|
|
#
?
Jun 17, 2014 23:12
|
|
|
Caconym posted:Time to yell at the vendor to optimize their drat SQL I think. Get a DBA on that, stat. You look like you're really heavily I/O boound (240k reads is ~1.8GB). A clustered index on partitions could make that much better. Whether there's an index at all. Or whether it's just a really, really big table. The idiocy of a query that requires 3 seconds of CPU time is another question entirely, but it's really impossible to tell what the cause of that is without an execution plan, the indexes on the tables used, how many tempdbs you have, etc. A DBA on in your org poking at it will be able to tell you.
|
|
#
?
Jun 17, 2014 23:14
|
|
|
deimos posted:Ahahahahahaha, good luck. Indeed. They're in the middle of a total reimplementation on more modern infrastructure, but they still support the old version. They really really hate messing with the 20 year old code responsoble for the current mess, but they can do it and this time they bloody well will.  evol262 posted:Get a DBA on that, stat. You look like you're really heavily I/O boound (240k reads is ~1.8GB). A clustered index on partitions could make that much better. Whether there's an index at all. Or whether it's just a really, really big table. The idiocy of a query that requires 3 seconds of CPU time is another question entirely, but it's really impossible to tell what the cause of that is without an execution plan, the indexes on the tables used, how many tempdbs you have, etc. A DBA on in your org poking at it will be able to tell you. The query is about 7 or 8 left outer joins on a smallish UNION in the centre with some CASE WHEN ELSEs and COALESCEs sprinkled in. Most of the tables involved have 3-4 million rows each. I've looked at the execution plan, there's a single index scan, the rest is all index seeks and lookups, all on clustred indexes with max 3.5% fragmentation and updated statistics. No single component of the query is resposible for more than about 12% of the cost. The DB is 15.8 GB in total, the server has 12 GB RAM allocated to the SQL Server buffer pool, all the IO should be RAM (we see no issues on disk I/O)
|
|
#
?
Jun 17, 2014 23:27
|
|
|
Caconym posted:The query is about 7 or 8 left outer joins on a smallish UNION in the centre with some CASE WHEN ELSEs and COALESCEs sprinkled in. Most of the tables involved have 3-4 million rows each. For a trifecta all you need is for it to be SQL Server 2012 and have a deadlock while rebuilding an index, that would be a LOVELY sight. There's a neat bug that corrupts indexes/data if there's a deadlock during parallel index rebuilds, exists in all versions of 2012 and wasn't fixed by SP2, only via hotfix. (link) deimos fucked around with this message at 23:35 on Jun 17, 2014 |
|
#
?
Jun 17, 2014 23:31
|
|
|
Riso posted:#yolo is the standard approach for small businesses. RAID 0 on the SBS server? CHECK
|
|
#
?
Jun 17, 2014 23:34
|
|
|
deimos posted:For a trifecta all you need is for it to be SQL Server 2012 and have a deadlock while rebuilding an index, that would be a LOVELY sight.  We have some deadlocks a day where an integration adapter tries to grab new data while it's still being hit with a stream of UPDATEs. (That's another thing, why the hell does a SELECT and an UPDATE lead to a drat deadlock? The SELECT should just wait dammit. At least it's always the SELECT that's killed.) SQL Server 2008 R2 (sp1 ) on this site thankfully, but I think I need to check with the DBAs tomorrow that that hotfix is installed on the one site we have on 2012...
|
|
#
?
Jun 17, 2014 23:42
|
|
|
|
| # ? May 21, 2024 15:59 |
|
|
Caconym posted:SQL Server 2008 R2 (sp1 The bug seems rare and requires a "perfect storm" type scenario.
|
|
#
?
Jun 17, 2014 23:48
|
|