I can't stop the company IT's power policy from being applied, but they can't stop me from having a script run every 30 mins to change it back.
Aioua: 1
IT: 0
What power policy is it that you have that they don't like? The only thing I enforce is disabling sleep/hibernate.
Just that. I disable sleep/hibernate, they don't like it.
Edit: rather, sleep isn't disabled, but setting it with no sleep timer. Always on!
Aioua on
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
They want it on sleep/hibernate? That's... odd. The only remotely rational explanation I can think of is to penny-pinch energy savings, but you'd have to have a shit ton of nodes for that to be worth the hassle of dealing with sleep/hibernation.
While I agree that being insensitive is an issue, so is being oversensitive.
They want it on sleep/hibernate? That's... odd. The only remotely rational explanation I can think of is to penny-pinch energy savings, but you'd have to have a shit ton of nodes for that to be worth the hassle of dealing with sleep/hibernation.
Yeah I think that's all it is. To be fair there are like a bajillion employees here.
This is what I get for using the standard build off of the WDS instead of rolling it myself.
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
0
Options
jaziekBad at everythingAnd mad about it.Registered Userregular
Found out about a bluescreen after the fact today, looked in the event logs and it gave me a stop code, which indicates an unhandled exception thrown by a program running in kernel mode, but no information as to what the application was. Nothing else in the system logs (apart from backup exec shitting itself on reboot. What a surprise).
I tried importing the memory dump into windbg, but since the server that crashed was windows 2000, and I'm using the most recent version of windbg, its throwing a load of errors about not having the correct dlls or symbols or something (I don't know I haven't used windbg much).
Does anybody know how I can get this setup working, i.e. using a debugger on a newer version of windows on a dump from a much older version)?
Found out about a bluescreen after the fact today, looked in the event logs and it gave me a stop code, which indicates an unhandled exception thrown by a program running in kernel mode, but no information as to what the application was. Nothing else in the system logs (apart from backup exec shitting itself on reboot. What a surprise).
I tried importing the memory dump into windbg, but since the server that crashed was windows 2000, and I'm using the most recent version of windbg, its throwing a load of errors about not having the correct dlls or symbols or something (I don't know I haven't used windbg much).
Does anybody know how I can get this setup working, i.e. using a debugger on a newer version of windows on a dump from a much older version)?
If it's complaining about symbols you might want to start here. You should be able to debug the 2k machine, as long as you get some 2k symbols on your debug host.
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
0
Options
TL DRNot at all confident in his reflexive opinions of thingsRegistered Userregular
website is inaccessible from client's office. Accessible from my PC here. DNS resolution seems fine. Already restarted the firewall.
I'm blanking here
Just because they're resolving doesn't mean they're reaching the site. It just means their DNS provider has an entry for the address. Ping/tracert the site to make sure traffic is getting there and back.
Do an NSLookup of the site from your PC, and put the IP address into the address bar of their PC instead of the URL of the site.
Make sure there's no entry for it in their hosts file, or that it's not on their restricted sites list.
Alright, so ping does not respond from our office or theirs, though our office can access the website.
Tracerts match until ~hop 15, at which point they diverge and the non-working location tries to hop to a different address (x.x.x.22 instead of .33). I'm waiting to hear back from the ISP.
0
Options
Apothe0sisHave you ever questioned the nature of your reality?Registered Userregular
website is inaccessible from client's office. Accessible from my PC here. DNS resolution seems fine. Already restarted the firewall.
I'm blanking here
Just because they're resolving doesn't mean they're reaching the site. It just means their DNS provider has an entry for the address. Ping/tracert the site to make sure traffic is getting there and back.
Do an NSLookup of the site from your PC, and put the IP address into the address bar of their PC instead of the URL of the site.
Make sure there's no entry for it in their hosts file, or that it's not on their restricted sites list.
Alright, so ping does not respond from our office or theirs, though our office can access the website.
Tracerts match until ~hop 15, at which point they diverge and the non-working location tries to hop to a different address (x.x.x.22 instead of .33). I'm waiting to hear back from the ISP.
Did someone just wiff the DNS entry?
Like, entered 22 instead of 33? If they're using split DNS then this is almost certainly a possibility? Or is this just a hop and there' said few more locations to go before reaching the endpoint?
R...routing?
What do you think this is, The New Yankee Workshop?
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
0
Options
AthenorBattle Hardened OptimistThe Skies of HiigaraRegistered Userregular
edited April 2014
Question:
How do you get buy-in for a change management process/change advisory board when another sysadmin only sees it as people second-guessing their work and slowing down things? Especially when the CAB doesn't meet on a schedule that matches how rapidly changes tend to go into the environment?
KakodaimonosCode fondlerHelping the 1% get richerRegistered Userregular
I wouldn't. It sounds like the CAB needs to meet more often or have a process in place to review things quicker. And on the IT side, you'll need to pull together better change plans and better scheduling.
And if the CAB doesn't report up the same chain that sysadmin does, it won't work.
Just to make sure this is the 100% ISP's problem, set your workstation to Google's primary & secondary DNS. 8.8.8.8, 8.8.4.4. Flush your dns cache and try the website. If it works, repeat at the customer's. If they still can't get the site to open and the tracert diverges, this is absolutely an ISP issue.
Ask their ISP if they made any recent changes for this client. I recently did some consulting work, and the customer's ISP made a change while the network guy was on vacation (and, like a goon, hadn't set an out of office response). They couldn't get an answer from him, so they just up and made some routing changes one day. Very similar results to this, where tracerts went really weird and unexpected directions. Had to get them on the phone to roll back the routing changes.
0
Options
DietarySupplementStill not approved by the FDADublin, OHRegistered Userregular
How do you get buy-in for a change management process/change advisory board when another sysadmin only sees it as people second-guessing their work and slowing down things? Especially when the CAB doesn't meet on a schedule that matches how rapidly changes tend to go into the environment?
Because shit happens. I don't care how good someone is at their job, we're all human, and mistakes get made, dependencies get overlooked, and someone may think a change is benign and months may go by, a discrepancy may be noticed, and no one has any record of the original change. Everyone should be accountable at their job, and system changes should be looked at and recorded.
Don't let people confuse CABs with deployments or releases, though. If admins can adhere to major system or architecture changes on a release schedule of their own, then a CAB isn't needed, but releases should be scheduled and documented (especially if there is going to be any sort of downtime, expected or otherwise). If something needs to change out of cycle, though, I would request a CAB of someone.
If people are looking at it as "they're second guessing my work" that's incredibly closed-minded of them. I would tell them that chances are, they're going to go to a shop someday where it's required. Better learn such practices now.
Edit, By way of anecdotal evidence: This happens all the time when a company acquires a much smaller company. When a group of 3 developers who built the entire enterprise at one company gets folded into the bosom of another larger entity, new practices and policies will always grate on them. Especially if they are subject matter experts and their administrative or production rights aren't stripped. They may make changes without anyone knowing. Sometimes these rights are as simple as process accounts that they have set up (which is good) but know the passwords to (which is bad) and changing them would break things. It's really a sea-change, and requires everyone to come to Jesus. It's for the better, no matter how much they want to resist.
How do you get buy-in for a change management process/change advisory board when another sysadmin only sees it as people second-guessing their work and slowing down things?
Depends on who's going to be on your CAB. If it's all IT staff, there's absolutely nothing wrong with peer reviews to production system changes. If you're confident that what you're doing is correct, it shouldn't bother you to have a peer review. If the CAB has people outside of IT, it's dumb. No one should rush into production changes. It's how shit gets broken.
Dev/test/lab? Go apeshit. Change whatever you want, whenever you want, as long as you know how to change it back if it doesn't work. Production change needs a system to tap the brakes and double check that "yes, this really is a good idea, it was tested, we have a plan in place to verify the success of the change once it's in production, and here's our roll-back plan in case something goes wrong".
Beyond simply "second guessing", a CAB process is also a way of documenting and organizing a record of the steps involved in a change and the steps required for verification and rollback. In any shop with more than about 3 people, this shit is important. Someone who has no idea what you did needs to know exactly where to look to see what's changed recently and how to undo it, if something blows up.
Especially when the CAB doesn't meet on a schedule that matches how rapidly changes tend to go into the environment?
How often does the CAB meet? Twice a week ought to cover most changes. Things need time to bake-in in your test environment anyway, and if you dont have a test environment that's all the more reason you need a CAB review. Also for emergency fixes, you have an emergency change process to do after-the-fact documentation.
CAB is a necessity in any grown-ass company, and this is coming from someone who utterly fucking despises the CAB process.
Cog on
+1
Options
AthenorBattle Hardened OptimistThe Skies of HiigaraRegistered Userregular
Oh, I'm 100% on board with a CAB. And the analogy of a merger/acquisition is probably the closest to what we have going on here. I'm just trying to figure out the right framework. Currently the CAB meets once a week. I think it's a single CAB that oversees all the apps we've placed on the "Fragile Artifacts" list, of which my officemate is one.
website is inaccessible from client's office. Accessible from my PC here. DNS resolution seems fine. Already restarted the firewall.
I'm blanking here
Just because they're resolving doesn't mean they're reaching the site. It just means their DNS provider has an entry for the address. Ping/tracert the site to make sure traffic is getting there and back.
Do an NSLookup of the site from your PC, and put the IP address into the address bar of their PC instead of the URL of the site.
Make sure there's no entry for it in their hosts file, or that it's not on their restricted sites list.
Alright, so ping does not respond from our office or theirs, though our office can access the website.
Tracerts match until ~hop 15, at which point they diverge and the non-working location tries to hop to a different address (x.x.x.22 instead of .33). I'm waiting to hear back from the ISP.
Is ping permitted inbound to their webserver (I don't permit ping inbound from the internet)? I suggested they telnet into the site (80) to see if they can see if service is up. It's pretty ghetto, but effective. "Telnet www.servename.com 80". If they cannot hit the service it'll just sit at "Connecting To url ..." and will fail. any other behavior means an http service responded. Most servers will pass you back a bad request error. www.microsoft.com is like a black hole though.
0
Options
AthenorBattle Hardened OptimistThe Skies of HiigaraRegistered Userregular
Ahahahaa.... hah.
I've just decided to take on something insane... Mapping out the integration points of our apps, and where data feeds back and forth.
Look, I don't give a shit if it's 2:30 on a Friday afternoon, fill the fucking coffee when you use the last of it you goddamn philistines. Some of us need that stuff.
Cause... we're tired.
Cause... .... we stay up late.
..... playing video games look fuck you just make the goddamn coffee.
My work's coffee may not be good, but it does come from machines that grind and brew one cup at a time, in less than a minute. I might never leave.
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Yeah, vac press is nice, but man that is a lot of fire hazard (unless you spend $texas on one of those fancy halogen heaters)
Mei, why the espro over a bodum press?
Yeah I mostly just do French press. Vac is just so fun. Making coffee with SCIENCE!
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Yeah, vac press is nice, but man that is a lot of fire hazard (unless you spend $texas on one of those fancy halogen heaters)
Mei, why the espro over a bodum press?
Unless Bodum produced a microfilter model while I wasn't looking, they just make regular french press right?
Espro uses microfilters to preserve the full body of the french press method and still have a cup free of sediments.
The sediments are good for you. Makes you grow hair on your chest.
Oh right sysadmin stuff.
Uh...
root@host:~# chmod 777 /dev/null
MIRITE?
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
0
Options
Apothe0sisHave you ever questioned the nature of your reality?Registered Userregular
The proper way to enjoy coffee at the office is to drink the shitty coffee swill they have in the communal germ-incubator coffee pots and then bitch about how fucking disgusting it is and the fact that you always seem to brew two pots every time you get one fucking cup and maybe someone else should pull their goddamn weight around here for once Cheryl and would it really kill the company to buy some liquid creamer instead of this powdered shit because this powder tastes like I'm creaming my coffee with Satans dandruff and even some of those little single serve creamer cups with the foil lid that never tears off in one piece would be fine even though its coffee creamer that sits un-chilled on a counter top for weeks at a time and even though it doesn't say you have to refrigerate it that still seems pretty fucking shady but man its still better than that powder by a mile am I right?
Posts
Just that. I disable sleep/hibernate, they don't like it.
Edit: rather, sleep isn't disabled, but setting it with no sleep timer. Always on!
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Yeah I think that's all it is. To be fair there are like a bajillion employees here.
This is what I get for using the standard build off of the WDS instead of rolling it myself.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
I tried importing the memory dump into windbg, but since the server that crashed was windows 2000, and I'm using the most recent version of windbg, its throwing a load of errors about not having the correct dlls or symbols or something (I don't know I haven't used windbg much).
Does anybody know how I can get this setup working, i.e. using a debugger on a newer version of windows on a dump from a much older version)?
If it's complaining about symbols you might want to start here. You should be able to debug the 2k machine, as long as you get some 2k symbols on your debug host.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Alright, so ping does not respond from our office or theirs, though our office can access the website.
Tracerts match until ~hop 15, at which point they diverge and the non-working location tries to hop to a different address (x.x.x.22 instead of .33). I'm waiting to hear back from the ISP.
Like, entered 22 instead of 33? If they're using split DNS then this is almost certainly a possibility? Or is this just a hop and there' said few more locations to go before reaching the endpoint?
PSN - sumowot
Because why would level 1 know what routing is
R...routing?
What do you think this is, The New Yankee Workshop?
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
How do you get buy-in for a change management process/change advisory board when another sysadmin only sees it as people second-guessing their work and slowing down things? Especially when the CAB doesn't meet on a schedule that matches how rapidly changes tend to go into the environment?
And if the CAB doesn't report up the same chain that sysadmin does, it won't work.
Just to make sure this is the 100% ISP's problem, set your workstation to Google's primary & secondary DNS. 8.8.8.8, 8.8.4.4. Flush your dns cache and try the website. If it works, repeat at the customer's. If they still can't get the site to open and the tracert diverges, this is absolutely an ISP issue.
Ask their ISP if they made any recent changes for this client. I recently did some consulting work, and the customer's ISP made a change while the network guy was on vacation (and, like a goon, hadn't set an out of office response). They couldn't get an answer from him, so they just up and made some routing changes one day. Very similar results to this, where tracerts went really weird and unexpected directions. Had to get them on the phone to roll back the routing changes.
Because shit happens. I don't care how good someone is at their job, we're all human, and mistakes get made, dependencies get overlooked, and someone may think a change is benign and months may go by, a discrepancy may be noticed, and no one has any record of the original change. Everyone should be accountable at their job, and system changes should be looked at and recorded.
Don't let people confuse CABs with deployments or releases, though. If admins can adhere to major system or architecture changes on a release schedule of their own, then a CAB isn't needed, but releases should be scheduled and documented (especially if there is going to be any sort of downtime, expected or otherwise). If something needs to change out of cycle, though, I would request a CAB of someone.
If people are looking at it as "they're second guessing my work" that's incredibly closed-minded of them. I would tell them that chances are, they're going to go to a shop someday where it's required. Better learn such practices now.
Edit, By way of anecdotal evidence: This happens all the time when a company acquires a much smaller company. When a group of 3 developers who built the entire enterprise at one company gets folded into the bosom of another larger entity, new practices and policies will always grate on them. Especially if they are subject matter experts and their administrative or production rights aren't stripped. They may make changes without anyone knowing. Sometimes these rights are as simple as process accounts that they have set up (which is good) but know the passwords to (which is bad) and changing them would break things. It's really a sea-change, and requires everyone to come to Jesus. It's for the better, no matter how much they want to resist.
Depends on who's going to be on your CAB. If it's all IT staff, there's absolutely nothing wrong with peer reviews to production system changes. If you're confident that what you're doing is correct, it shouldn't bother you to have a peer review. If the CAB has people outside of IT, it's dumb. No one should rush into production changes. It's how shit gets broken.
Dev/test/lab? Go apeshit. Change whatever you want, whenever you want, as long as you know how to change it back if it doesn't work. Production change needs a system to tap the brakes and double check that "yes, this really is a good idea, it was tested, we have a plan in place to verify the success of the change once it's in production, and here's our roll-back plan in case something goes wrong".
Beyond simply "second guessing", a CAB process is also a way of documenting and organizing a record of the steps involved in a change and the steps required for verification and rollback. In any shop with more than about 3 people, this shit is important. Someone who has no idea what you did needs to know exactly where to look to see what's changed recently and how to undo it, if something blows up.
How often does the CAB meet? Twice a week ought to cover most changes. Things need time to bake-in in your test environment anyway, and if you dont have a test environment that's all the more reason you need a CAB review. Also for emergency fixes, you have an emergency change process to do after-the-fact documentation.
CAB is a necessity in any grown-ass company, and this is coming from someone who utterly fucking despises the CAB process.
Is ping permitted inbound to their webserver (I don't permit ping inbound from the internet)? I suggested they telnet into the site (80) to see if they can see if service is up. It's pretty ghetto, but effective. "Telnet www.servename.com 80". If they cannot hit the service it'll just sit at "Connecting To url ..." and will fail. any other behavior means an http service responded. Most servers will pass you back a bad request error. www.microsoft.com is like a black hole though.
I've just decided to take on something insane... Mapping out the integration points of our apps, and where data feeds back and forth.
Cause... we're tired.
Cause... .... we stay up late.
..... playing video games look fuck you just make the goddamn coffee.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Here the coffee is not good... but holy fucking shit, is it loaded.
But my set up is a little...uh...overboard.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Mei, why the espro over a bodum press?
Yeah I mostly just do French press. Vac is just so fun. Making coffee with SCIENCE!
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Espro uses microfilters to preserve the full body of the french press method and still have a cup free of sediments.
Can't you like... pretend that users are unable to work the coffee machine or something so I can enjoy the posts too?
The sediments are good for you. Makes you grow hair on your chest.
Oh right sysadmin stuff.
Uh... MIRITE?
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
The proper way to enjoy coffee at the office is to drink the shitty coffee swill they have in the communal germ-incubator coffee pots and then bitch about how fucking disgusting it is and the fact that you always seem to brew two pots every time you get one fucking cup and maybe someone else should pull their goddamn weight around here for once Cheryl and would it really kill the company to buy some liquid creamer instead of this powdered shit because this powder tastes like I'm creaming my coffee with Satans dandruff and even some of those little single serve creamer cups with the foil lid that never tears off in one piece would be fine even though its coffee creamer that sits un-chilled on a counter top for weeks at a time and even though it doesn't say you have to refrigerate it that still seems pretty fucking shady but man its still better than that powder by a mile am I right?
....
But i mean..
Your ways are ok too I guess.