|
less than three posted:Doesn't help by quoting it. If his Larry/Lewis is anything like my Larry/Lewis there is a probability approaching zero that he will ever know his first name escaped here.
|
#
?
Oct 17, 2014 05:05
|
|
|
|
| # ? May 25, 2024 23:37 |
|
|
Maybe he's supporting a terrible college football coach.
|
|
#
?
Oct 17, 2014 05:36
|
|
|
Details of our IT reorg have started to leak, and it sounds like a lot of clowns will be moving around the ring, but the circus isn't changing. The current IT manager (Chuckles) will now just be managing 3 departments and will be the "face" of IT to our users. Which is fantastic as hardly anyone believes a word that comes out of his mouth. Since we always need more managers, a existing manager (Carl) is being moved up a tier, and now my existing manager (Goober) will report to him instead of the previous guy. This has created some tension, as Carl was a peer to Goober and has been with the company less then a year. Also him and Goober don't get along very well. I have only briefly met Carl and while he seems a ok person and the people that report to him currently seem to be ok with him, I have no idea what he'll do about us. No word of they plan on hiring more people to actually do the work, if we'll just keep adding managers.
|
|
#
?
Oct 17, 2014 06:08
|
|
|
less than three posted:Doesn't help by quoting it. it does if you're on LinkedIn. Well, it will when I'm a hiring manager again.
|
|
#
?
Oct 17, 2014 06:27
|
|
|
Exit Strategy posted:Same guy as the "can you change your hours" ticket: Why have you not fired the customer? 
|
|
#
?
Oct 17, 2014 13:26
|
|
|
Get in this morning and not even had enough time for the morning haze to lift from my eyes before a Tech Arch is bitching to me about a Project Manager while the Project Manager is bitching about the Tech Arch at the same time in two different lync windows. I am not your babysitter children.
|
|
#
?
Oct 17, 2014 13:38
|
|
|
I know this is probably an unpopular opinion with the *nix neckbeards around here, but I've come to the conclusion Nagios is goddamned terrible.
|
|
#
?
Oct 17, 2014 13:42
|
|
|
flosofl posted:Did you slip up in the midst of your rage and give us his real name, there? B-b-b-but muh OPSEC!
|
|
#
?
Oct 17, 2014 14:35
|
|
|
psydude posted:I know this is probably an unpopular opinion with the *nix neckbeards around here, but I've come to the conclusion Nagios is goddamned terrible.
|
|
#
?
Oct 17, 2014 14:37
|
|
|
Give Zabbix a shot.
|
|
#
?
Oct 17, 2014 14:40
|
|
|
BigPaddy posted:Get in this morning and not even had enough time for the morning haze to lift from my eyes before a Tech Arch is bitching to me about a Project Manager while the Project Manager is bitching about the Tech Arch at the same time in two different lync windows.
|
|
#
?
Oct 17, 2014 16:06
|
|
|
BigPaddy posted:Get in this morning and not even had enough time for the morning haze to lift from my eyes before a Tech Arch is bitching to me about a Project Manager while the Project Manager is bitching about the Tech Arch at the same time in two different lync windows.
|
|
#
?
Oct 17, 2014 16:15
|
|
|
I mostly just use Lync to poo poo on people.
|
|
#
?
Oct 17, 2014 16:23
|
|
|
I like SCOM 
|
|
#
?
Oct 17, 2014 16:26
|
|
|
flosofl posted:If his Larry/Lewis is anything like my Larry/Lewis there is a probability approaching zero that he will ever know his first name escaped here. I changed the name I was using for him, and didn't just cmd-R the post in a text editor. Volmarias posted:Why have you not fired the customer? Because it's not worth firing him? He'll quit, eventually, when he realizes we're not a free sysadmin service. I've seen it before. This time, he had the forethought to submit the ticket through a different email address, successfully tricking us into talking to him after the legal-threat poo poo. So I flagged his license key with the instructions for Nuisance Mode. It could be worse, he could be the user whose autoreply doesn't read "Thank you for submitting your ticket," etc when they send one in. Instead, it reads "Hello, [user]. Please ensure that your invoices are paid before presenting a support ticket for consideration. Reply to this email once paid and we will begin reviewing your issue. Exit Strategy fucked around with this message at 16:34 on Oct 17, 2014 |
|
#
?
Oct 17, 2014 16:30
|
|
|
psydude posted:I know this is probably an unpopular opinion with the *nix neckbeards around here, but I've come to the conclusion Nagios is goddamned terrible. We all know it sucks, too. And trying to have multiple masters is even worse. That's why opsview and other "make it less painful" tools exist. Zabbix is better if you need plug and play. But the hot poo poo these days is collectd monitoring (or similar) feeding into a dashboard and flexibly expanding infrastructure depending on load. Kill problem children and rebuild them.   Seriously use another tool, though
|
|
#
?
Oct 17, 2014 16:32
|
|
|
We'll call him L Simpson. No wait, that's too obvious. Lisa S!
|
|
#
?
Oct 17, 2014 17:04
|
|
|
psydude posted:I know this is probably an unpopular opinion with the *nix neckbeards around here, but I've come to the conclusion Nagios is goddamned terrible. Nagios is unadulterated poo poo and I don't understand its popularity. Trying to get it to pull and display Windows event logs in an easily-parsed format is a nightmare, and I never successfully did it. I followed the exact syntax in the nagios manual for WinEventLog and it didn't output at all the way it was written that it would. Endless experimentation after doing it by the book got me an eighth of the way to where I needed to be. The flipside is that we have like five other tools for monitoring and none of them are properly configured and we end up getting 20 different email/text alerts when server response times are abnormal. And none of them are looking at Windows event logs.
|
|
#
?
Oct 17, 2014 18:20
|
|
|
We had a laptop with CryptoLocker come in. We discussed what to do with it, and we decided we'd plug it into one of our Verizon hotspots and see if we can use the decryptolocker site to pull the encryption keys and get his files backed up. One of my coworkers grabbed it first, so he loudly makes sure the wifi switch is turned off before turning it on, and boots into Windows normally. He logs in as the user and starts opening spreadsheets to read the Excel error message. Like after ten minutes of this, he turns to me and asks why the hard drive light would be flashing so much.
|
|
#
?
Oct 17, 2014 18:22
|
|
|
evol262 posted:We all know it sucks, too. And trying to have multiple masters is even worse. That's why opsview and other "make it less painful" tools exist. My boss is all about doing things the old way even if it's more work. Like a lot of our poo poo is much more easily done via the UI, but he perfers the CLI. My stance on the matter is to go with whatever is the most quick/efficient. Anyway, at my last job we used WhatsUpGold, which isn't perfect (nor is it free, but what's 2 $3000 licenses when the poo poo we buy routinely costs six figures?), but is a hell of a lot easier to manage than Nagios.
|
|
#
?
Oct 17, 2014 18:46
|
|
|
anthonypants posted:We had a laptop with CryptoLocker come in. We discussed what to do with it, and we decided we'd plug it into one of our Verizon hotspots and see if we can use the decryptolocker site to pull the encryption keys and get his files backed up. One of my coworkers grabbed it first, so he loudly makes sure the wifi switch is turned off before turning it on, and boots into Windows normally. He logs in as the user and starts opening spreadsheets to read the Excel error message. Is there a site like this for cryptowall yet can't seem to find it.
|
|
#
?
Oct 17, 2014 19:13
|
|
|
So I know many of you hate BackupExec (I've never dealt with it luckily). Here's your chance to talk to the developers: https://www.reddit.com/r/IAmA/comments/2jj5lx/we_are_the_backup_exec_development_team_ask_meus/
|
|
#
?
Oct 17, 2014 19:34
|
|
|
socialsecurity posted:Is there a site like this for cryptowall yet can't seem to find it. 
|
|
#
?
Oct 17, 2014 19:36
|
|
|
What's pissing me off, My company is run by a bunch of idiots who have created a corporate culture where only "Yes" (wo)men are promoted, and anyone who may warn that some idea coming down from on high (directors or one of the 5 layers of VPs we have) may be less than a stellar idea is labeled as having a negative attitude finds themselves relegated to grunt work. The most recent bit of idiocy is the corporate mandate to have all company applications automatically log users off and force re-authentication after 15 minutes of inactivity. Now the nature of the job and the lack of decent tools support means that we use multiple applications in the course of our duties and it's not uncommon for us to be away from any given tool for more than 15 minutes. Meaning that we are regularly blessed with a message telling us we have to log back in. The stupidity here is multi-faceted. First the tool that's generating the most annoyance doesn't remember its state when it logs you off meaning that after we click through to get back in we have to navigate back to where the information we need is. Now note what I said about clicking through? that's because these are web-based tools and the browsers cache log in credentials which means that their 15 minutes log out timer does gently caress all for adding additional security since anyone who can get access to your desktop will be able to click through the authentication prompts with precious little more effort than if they never existed at all. No loving sense what so ever. And this is pretty typical for the bone headed ideas that are constantly rolling down. What pisses me off the most is that somehow these morons make more money than me, a poo poo ton more. They do poo poo like this all the time then cry about how we aren't meeting whatever arbitrary metric they've decided should be focused on for.... reasons. My business unit is a Network Operations Center yet management insist on treating us like a Call Center (They take calls they must be a call center yuk yuk yuk). Meaning that instead generating metrics that a NOC should be concerned with, down elements, mttr, network availability; they generate metrics that a call center would, Speed to answer, call volumes, ticket churn. The result being that we're an extremely dysfunctional organization that is constantly playing whack-a-mole and our work prioritization is based on the squeaky wheel principle. Since the focus is on call volumes they've prioritized taking calls the result of which is that we end up with higher call volumes than we'd have if we actually behaved like a NOC because issues only get addressed when someone bothers to call in and complain. Instead of encouraging our technicians to be thorough we're focusing on speed of call handling meaning that simple stuff is getting missed, customers are getting poor service because technicians who are trying to make those arbitrary metrics management has dreamed up and they're punting or blowing off customers rather than taking the time to make sure they've really done their due diligence on a particular issue. To aid in measuring their metrics they've chosen to focus on they've added a legion of little book keeping tasks that exist for no reason other than to generate metrics (apparently they're unfamiliar with the concept of observer effect. And thanks to the lack of uniformity in processes people regularly miss those steps because there is no way any human can be expected to remember the sheer volume of crap they make us keep track of. As if we don't have enough to know when it comes to actually doing the Network Operations part of our loving jobs. Oh and did I mention that the metrics they are using are fundamentally flawed because they haven't built in the capabilities to our systems to be able to properly measure what they want to measure? For example one of the metrics we are measured on for our reviews (read raises) is ticket touches. The problem is that the field in the crappy ticketing system that they rammed down our throats that would make it possible to accurately measure that metric (date/time stamp for ticket interactions) isn't propagated to the database they use to generate their reports. And thanks to the yes men culture that exists instead of the people generating said reports telling their bosses "we can't do that because of this, can we fix this?" so that maybe just maybe actions could be taken to rectify the deficiency they just cobble together some hack'd up approximation and pretend it's actually a useful metric. And in keeping with modern American corporate culture of maximizing profits in the short term by not investing in the business (might hurt the stock price if they reported smaller profits after all) they don't provide us with the resources to keep up with the increased work load they continue to dump on us. They don't hire enough bodies they don't invest in effective tools, they don't listen to our feedback on how things could be improved. The sales and marketing people are constantly making commitments that we simply lack the technical capability and head count to dream of meeting. They keep piling more and more work on our already overused workforce all while pushing down different snowflake procedure's that have to be followed in some circumstances but don't apply in others. One would suppose that at some point they might see the results of this but going back to yes men culture and managers pay being contingent on making these arbitrary goals they massage and manipulate the metrics to present a distorted version of reality that cloaks the truly appalling state of the companies operations. Finally in the ultimate bit of clueless irony the same executives who have engendered this "yes (wo)man" corporate culture through their promotion and hiring practices and pushed down all this crap that makes doing the job so much more difficult than it needs to be are promoting this idea that we need to provide an "effortless customer service" experience. Oh yeah and they expect to do this without having to actually spend any additional money. Meritocracy my rear end.
|
|
#
?
Oct 17, 2014 22:21
|
|
|
Thanks internet explorer, I did want to search Bing for this internal address instead of just connecting to it, you're right I should have added http :// first. How foolish of me.
|
|
#
?
Oct 17, 2014 22:30
|
|
|
Every single MSP in the history of forever.
|
|
#
?
Oct 17, 2014 22:30
|
|
|
CitizenKain posted:We had a port scan/intrustion test done on a important vlan on day by some auditors with the help of the security person. Whatever they did caused an old IDS to pretty much poo poo itself and it failed closed instead of open. That was a fun day. Reminds me of the time we had to start providing SLA compensations on a new vpn tunnel product we were rolling out. In the meeting with the software guys I'm asking how are we going to be able to document what failures are our fault and what ones we can exclude and suggest that we should generate tickets to track that, they no lie, with a straight face told me "oh this technology is solid it's not going to cause any problems." I was like are you serious? How long have you been in this industry? nothing ever goes right the first time. Management thought I was over reacting. 2 months later after having to negotiate out of a couple hundred grand in missed SLA fines we started generating tickets to track issues with vpn tunnels.
|
|
#
?
Oct 17, 2014 22:37
|
|
|
Manslaughter posted:Thanks internet explorer, I did want to search Bing for this internal address instead of just connecting to it, you're right I should have added http :// first. How foolish of me. Firefox, but with localhost. Also adding http:// does nothing. Also changing the flag does nothing. Have some n directory length localhost url? I hope you like the google page you're going to see telling you it can't find anything. Extending to chrome ignoring cookies set to localhost, so you need to change it to local.host in the hosts file to work correctly without loving around. Firefox to an extent as well, but it tends to play nicer. I really don't understand why browsers handle localhost so differently. Then again a lot of browser differences are stupid. gently caress web development (and PHP, but that's a given).
|
|
#
?
Oct 18, 2014 03:39
|
|
|
I've had the opposite problem with Chome and this happened TODAY. I have a day job but also volunteer for IT for the active chapter of my Fraternity. I went to check because they got a warning from their ISP about torrenting. So I check and see no major downloads, but still a lot of traffic from a particular IP. Like and idiot instead of doing a reverse DNS I just flat put it into Google CHrome - which due to the Omnibar instead of searching attempted to go straight to the IP. This hit the WebFilter for pornography. Forced a Google search to find the IP was for ImageFAP (no pictures, just links - thank god). Way to make my heart skip a beat.
|
|
#
?
Oct 18, 2014 04:04
|
|
|
It pisses me off knowing every time I fix a large scale issue that none of these other morons I work with have the talent to figure out, I'm met with torches and pitchforks asking me how and why this happened to start with... like it was my fault your stupid system for douche bags was down to start with. You send me an elegant loving thank you card, then gently caress off and die forever.
|
|
#
?
Oct 18, 2014 04:18
|
|
|
Is your NOC taking calls from clients? Do you have a helpdesk?
|
|
#
?
Oct 18, 2014 04:19
|
|
|
Chalets the Baka posted:Nagios is unadulterated poo poo and I don't understand its popularity. Trying to get it to pull and display Windows event logs in an easily-parsed format is a nightmare, and I never successfully did it. I followed the exact syntax in the nagios manual for WinEventLog and it didn't output at all the way it was written that it would. Endless experimentation after doing it by the book got me an eighth of the way to where I needed to be. Nagios isn't *that* bad. It's kind of clunky, it certainly has its quirks and annoying bugs/design decisions, and you definitely need to have a third party front end if you're monitoring more than a few systems, but it works just fine if you take the time to design and configure your monitoring platform properly. My company is currently using Nagios with a Centreon front-end to monitor about 1600 hosts with 15k service checks. Takes a dozen pollers to handle that many checks, and the database back-end is pretty I/O-heavy, but it works quite well overall, it's free, and the flexibility is a huge plus. We constantly have business units coming to us asking "can you monitor X?" where X is some really obscure statistic or condition, and the answer is almost always "yes", since if there isn't a plugin out there that can do it already, I can usually hack together a script that will handle it in an hour or so (well, maybe a couple hours if it's some Windows crap since I don't really know any Windows scripting languages...  ). Seriously though, if one of the included plugins isn't doing what you need, just find an alternative out there, or roll your own. The API is dead simple; all you have to do is return a particular exit code for the check condition and whatever text you want describing the check result (plus some optional and easily formatted performance data if appropriate). If you know enough PowerShell black magic to parse the event log, it should be pretty simple to write a quick script that will check for whatever entries you need to alert on (though to be honest, I'm not sure what trouble you'd be having with the default plugin, as we're using it to alert on specific event log entries on several of our systems without a problem).
|
|
#
?
Oct 18, 2014 04:53
|
|
|
edit: crap, wrong thread
RFC2324 fucked around with this message at 10:29 on Oct 18, 2014 |
|
#
?
Oct 18, 2014 08:07
|
|
|
RFC2324 posted:This just came up on my FB. Nice story, but http://forums.somethingawful.com/showthread.php?threadid=3186581
|
|
#
?
Oct 18, 2014 10:07
|
|
|
Dr. Arbitrary posted:That'd fit great in this thread: Yeah, was supposed to be the stdh.txt thread.
|
|
#
?
Oct 18, 2014 10:29
|
|
|
lampey posted:Is your NOC taking calls from clients? Do you have a helpdesk? Yes we take calls from Clients, since we're so dysfunctional we get far more inbound calls that we should, ideally we would be primarily making outbound calls to report problems and get the help of onsite staff to assist in layer 1 troubleshooting, but thanks to chronic under-staffing and poor process planning on the part of our management we spend most of our time taking inbound calls as our clients notice and report problems with the network infrastructure we support (squeaky wheel work prioritization). It primarily comes down to a question of what paradigm being applied to the operation. Since management is applying a Call Center paradigm to the NOC they end up focused on metrics that a Call Center would. To give an example of the detrimental affect this mindset has, one of the biggest problem we've faced as an organization for as long as I've worked here is information overload and being able to cut through the chaff to find the actual substance, one of the biggest contributors to that is how our ticketing system works and how automation has been applied to it. I spent a number of years doing reporting for the organization, so I've gained a really good understanding of where these problems are. Under our current system when a device goes down our monitoring system generates an alarm when this happens a second system (rubyrules) sees this alarm and then generates a ticket in our ticketing system. When the device comes back up Ruby then tells the ticketing system that the event has ended and moves the ticket into a ready to close state. A chron job on the ticketing database then looks for tickets that have been in a ready to close state for longer than the target time (I'm not sure what the time is now but originally it was 24 hours) When this was first implemented we had requested that the ticket remain in that ready to close state for 24 hours and only close if the event did not recur in that time period. Within the first couple months I noticed that the ticket volumes were just insane, so I got the ruby programer to send me over his rules and when I deciphered the logic of his processes I found that instead of reactivating the previous ticket when an additional event occurred it was simply creating a new ticket in addition to the original. Now understand when I ran the numbers on this stuff I found 90% of our ticket volume was on device events that took place within 72 hours of a previous event. So I recommended that we correct the automation so that tickets would remain in a ready to close state for 72 hours before closing and for new events to be appended to the existing ticket and that ticket be put back into an active state when that occurred which would instantly reduce our ticket volume by reducing duplicate tickets for the same issue, we would then be able to separate the wheat from the chaff by looking at ticket age so the oldest tickets would be on chronic issues and we could then concentrate our efforts on resolving those while ignoring the short 1 time stuff that tends to clutter up the process. Instead of implementing what was a relatively simple straight forward solution to our problem they ended up implementing a very complicated set of ruby rules that would try to track repeat events and manage tickets that way. Our ruby rules exploded from a few hundred to several thousand and the end result was an extremely complicate process that did nothing to reduce our ticket volume and that was so complex that it's constantly having unintended behavior due primarily to the fact that it's so damned complicated that no one can actually analyze it's logic processes to figure out where it's screwing up. Now when I proposed my far more simple and elegant solution to the problem I was told that they weren't interested because it would increase the reported MTTR and that reducing ticket volume would make it harder to make the case for more headcount. Because I guess all these highly educated directors and VPs are too stupid to understand that metrics are not absolute. Never mind that the MTTRs they are reporting now are essentially fabrications that bear no relation to reality. It would look worse so we can't do it. Never mind that reducing the volume of junk tickets would result in more efficient and effective operations by letting us focus our limited resources on real problems. Remember it wasn't because they thought it was a bad idea or that it wouldn't work. Their objection was that they thought it would work and that it would make them look worse because the numbers reported would change and my appeals that it was simply a matter or explaining the situation to upper management and managing their expectations fell on deaf ears. So we remain in a dysfunctional state primarily because of what metrics management has decided to focus on rather than on what would result in the most efficient and effective operations. This is just one example of how applying the wrong paradigm to an organization can have adverse affects to it's operation. It's not like their way of doing things is actually resulting in issues being resolved faster, it just looks like it because of how they are measuring. Kind of like how giving a measurement in centimeters results in a larger number than giving the measurement in inches would.
|
|
#
?
Oct 19, 2014 17:20
|
|
|
Apparently I have to go to a two-day long meeting, that is 3 hours drive away from where I live even when travelling off-peak. To get there I'll have to drive past Birmingham (second largest city in the UK) and then down an hour's worth of local roads through schools and housing estates. Oh, and it starts at 10:30AM, finishes at 5, then resumes at 9 the following morning to finish at 4. And the hotel they booked me is in a different town, Walsall, which is bad enough on its own but it's at the far end of a road which is basically a car park for 2 hours in the morning and 3 hours in the evening.
|
|
#
?
Oct 19, 2014 20:36
|
|
|
Scaramouche posted:Stay safe Drupal ghosts: We're a Drupal campus. Between that and a CAS vulnerability, this week has been funnnnnnn
|
|
#
?
Oct 19, 2014 21:51
|
|
|
poo poo that's pissing me off: Fracturing my finger so I can't use that one to type.
|
|
#
?
Oct 19, 2014 23:35
|
|
|
|
| # ? May 25, 2024 23:37 |
|
|
dogstile posted:poo poo that's pissing me off: Fracturing my finger so I can't use that one to type. And you know you're going to use it anyway. You're going to use it every 90 seconds. You know.
|
|
#
?
Oct 20, 2014 02:02
|
|