Re: Best traps
@abelsoul: There have been some studies done that suggest that cats do this in order to teach us Terrible Big Things how to Cat properly and how to hunt. :)
393 posts • joined 16 Jul 2007
@abelsoul: There have been some studies done that suggest that cats do this in order to teach us Terrible Big Things how to Cat properly and how to hunt. :)
There was also the time that I was stuck repairing an eMachine (A terrible, low end, cheap machine that (barely) ran windows 98se, let alone XP which some ignoramus slapped on it) which had a _destroyed_ USB port.
Apparently, someone tried plugging in a USB device on the back port, and it wouldn't fit. Not content with turning the USB cable over, they just rammed the thing in, breaking the connector in such a way that it shorted the portion of the chip the circuit traces ran to.
There was exactly *two* usb ports on the entire machine to begin with- one in front, one in the back. (the one in the back was the trashed one).
Ironicly enough, the thing still had the display sticker plastered on the front of the case proudly proclaiming that would "never need to be upgraded!". (Primarily because you _couldn't_ upgrade it, and they were cheap enough that you tossed it and bought a Real Computer afterwards.)
The BOFH (and the PFY) have actually changed companies at least once since then.
But it's a possibility.
... I have your backup here on my desk, actually.
(One of our less clueful tape monkeys actually did put an LTO1 barcode on a cleaning tape. Not surprisingly, it never worked, and it confused the hell out of the backup software.)
... unless you have an IT staff that constantly monitors it and keeps on top of the filtering configuration.
Granted, the one I'm in charge of is at $work; it's partly to keep the amount of pink off the monitors, but also to block failbook and other social media sites from the masses to keep productivity at acceptable rates. Finally, it also acts as a reasonable malware blocker.
Ironicly, the one they put in charge of the nanny filter is the one who is opposed to censorship on almost a religious level; but I understand and accept the business's reasons for it.
IIRC, one of the porn sites is _actually doing this._ Can't remember which one, and can't access it because I'm at work (which frowns upon that sort of thing)
I can say that he's not, but he's popular enough with the small group of voters who come out en masse to re-elect him every time, if only because no one else will take the time to actually vote.
(that, and there's been a good amount of 'legal but not ethical' election tampering here too by mucking around with the polling places and other shenanigans.)
Same here; The hardware is actually pretty decent as well and is pretty robust for the price.
I'm also hoping that HPE doesn't crank up the support pricing and reduce the level of said support; Nimble's support has been extremely good for the times we've needed it outside of replacement part swaps.
Nimble's official press release:
and a blurb by Rueters:
Price is 1 Billion USD. (That's $1,000,000,000 USD for the nitpickers in the crowd)
That was... unexpected. I just hope that the service and support doesn't change; They've been extremely good about getting replacement parts to us when stuff breaks, and their analytics are very useful as well- we've used their data to prove to one of our vendors a couple times now that the problems with their application are not storage or IO related, at least not at the appliance or hypervisor level.
Sitting on my desk right now is a 'cheap' (~$80 USD) car jump battery pack sold by an import tool company infamous in the US for 'cheap' tools that are good in a pinch, but have widely variable quality control standards.
It's using a Lithium iron phosphate (LiFePO4) battery pack, and I just jump-started a 4 cylinder ford ranger truck with it. It's not much larger than a hardcover novel in it's travel case, and it works quite nicely. It's also much lighter then the older style portable jump kits that use a lead-acid battery.
Even five years ago, something like this either didn't exist, or was four or five times the cost. I love living in the future.
Well, there was no mention of feeding them into a tree chipper or some other permanent life process termination device? :)
(Yes, poor taste. I know, I'll show myself out.)
for managing the various systems in the environment, and especially if the company's already blown the wad for System center as opposed to something like Solarwinds Orion or any other number of server management application suites.
SCOM is obviously not perfect- it's a pain in the butt to install, configure, and operate if you've never used it or had good training on it, there's a snootload of 'legacy code' in it from the old SMS and MOM days, and overall it's a shambling, baroque monstrosity.
I've had a few supervisors at the current place I work at who are pretty lax about slacking off- as long as:
a) the work gets done
b) it's not *too* invasive
for example, on super slow days (holidays, primarily, where I was required to be there for break/fix, but since things were running well I ended up having very little to actually *do*) I'd have my personal laptop in and was playing movies, offline games, or using the company's guest wireless to play minecraft or some such.
And yeah, I could *probably* write off El Reg as "technical research", but meh. (as I type, I'm waiting for an content index rebuild on a pair of DAG nodes, and keeping a weather eye on the backup job that's running, and waiting for a vendor to ping me.
The CS300 and earlier models in that class, it's iSCSI over 10Gb Ethernet. the CS500 has FC as an option, and beyond that I can't answer to, not having one sitting in my data center. :)
Beyond that, I have no idea how FC works on the nimble platform, as we have been using it as iSCSI only. Our CS500s don't have the FC option on them.
Things it can't do (and probably never will):
CIFS/SMB, aka windows file sharing. You can stand up a machine running your favorite OS with an SMB server, and back-end the storage off a nimble, but you can't run it directly from the appliance.
NFS: Essentially the same thing; Nimble has a whitepaper on how to set up a pair of linux based machines to create an NFS gateway with the array ac ting as the backend storage.
I could potentially understand if an a disk shelf's IO controller went out and failed in a way that corrupted the data being written to the disks. The likelihood of that happening is.... Well, it's a large-ish number to one against. I can dig the IO controller rolling over, I've seen that a few times; the backup took over straight away, and the machine had an alarm light on it until the blown controller was swapped for a good one.
(In the ~20 years I've been in the industry, I've seen two raid controllers blow themselves up; one did in fact take the array with it, the other failed in a manner that rendered the battery backed cache non-usable, which only crippled system performance.)
At one point, I had a rubber chicken in my tool kit, just in case I needed to tap some of the deep magic for fixing the various and sundry issues I've run across. (this was back when you could reliably recover from a machine getting a virus by removing the damaged files and putting known good copies back in; no such luck now.)
There was also the 'incident' with me performing a data recovery off a failing drive by wrapping it up in a towel and ice pack to keep the controller cool enough to perform an emergency disk clone to a known good drive- I managed that neat trick exactly twice.
Bring back the Hot dog theme.
Or at least give us a way to re-color the various elements.
It's friday, I need one.
"Just wait until the Chines buy big in Vendor/manufacturing land......"
(aka all of IBM's x86 products: laptops, desktops, and servers)
Apple and others (via Foxconn)
Unless you are trolling?
Possibly make the body of the telepresence unit slightly more customizable? (i.e., the ability to change the color of the unit via LEDs, etc.) possibly the ability to swap the outer shell might be cool; (i.e., for a sci-fi themed convention, put dalek shells on them, or some such.)
Bloody hell yes. requiring me to have a (very specific) java RE in order to configure the damn Jetdirect, and then on top of it the applet fails better than 2/3rds of the time? Absolute FAIL.
That's partly why most of the older ones were configured via Telnet- loads faster, especially if you knew the (not-so) secret method for getting it to pick up an IP address of your choosing. (i.e., connect the jet direct to the same subnet as your workstation, manually add the MAC address to your ARP table with the address, and then ping said address.)
Those were also mostly external Jetdirects, I should mention. Internal ones (generally) were slightly easier to use in that regard.
(Re: the "Code in..." stunt)
I see what you did there.
I used to get random calls on my cellphone from such amusing places as south africa and surrounding countries. I pulled that stunt a few times after the 3rd or 4th caller to do that.
Sadly, I tired even of that and changed my number a month or two later.
... But this one might earn a 'dumbest hacker EVER' award.
Amusingly enough, one of the places I used to do service for was a commercial laundry. The computers on the inbound processing line where the cheapest machines we could get, and they were mounted on a 18 inch high stand, because they would have rusted to the floor otherwise when it was power-washed every night. (The machines still were nasty gross things that the techs hated to PM because odds were you'd need to change clothes afterwards. Ugh!)
I tend to *invent* new Words of Power when I'm trying to resurrect failed equipment, usually because the CIO (or worse, the entire company) is beating down my door wondering why their email isn't going through... :)
Mine's the one with the journal with black covers titled 'An Incomplete Dictionary of the Obscene and Profane' in the front pocket
For a while, our VMware stacks at $work were all Opteron 6xxx series- VMware charges by the socket for the hypervisor, and cramming 48 cores worth of CPU on a 4 socket box was very attactive for us. the standalone machines (very few and far between) were intel based, though.
at this point though, we are moving to intel for it's replacement (E5-26xx v3), because when I looked last year there were exactly *two* companies selling AMD boxes, and they were all 2 processor generations old. We have a BI application that that used to take 40 minutes to process on the AMD hardware; once we migrated it to the new stuff the processing time dropped by 60% easily.
If AMD can crank out server chips again and get vendors to actually make boxes, I'll cheerfully recommend them- they were certainly cheaper then the intel equivalent at the time.
Hell, it's the first thing they (used) to teach new hires here:
Job one is 'get paid'.
a couple years ago we changed the system we used for tip reporting. the masses almost set the system analyst on fire due to the fact that the changes to how tips were cashed out were not communicated adequately. Fun times.
the PT was more of a microvan than a minivan; You could put quite a bit into the thing if you folded up the rear seats (or removed them entirely, which was easily done via a couple latches), and it did have a unique styling all it's own. (It took the styling cues from the Prowler, which took it's styling cues from the hot rods of the 30's)
the PT was not without it's faults- it had the same powertrain as the Neon platform (with all the quirks and problems), the turning radius sucked, and the gas mileage was nothing to write home about. It also had (at least out here in the desert) of chewing through it's battery almost every year.
I was wondering when Project Binky would make an appearance in this thread.
the entire thing (so far) is well worth a watch. It'll be interesting to see how it looks and drives when it's done.
IIRc, it was one of the other commentards that was complaining about password change replication.
I've *never* need password changes not propagate within 3-5 minutes at the absolute outset, and usually it's because the end user was hammering on the local DC after the support droid at the site on the far side of the network reset the user's password on their local DC, with the end result of the user's account still getting locked out at the PDCe at the hub site between the two.
New device joins, coming in via a MDM out in the cloud via federation or proxy or some other means? Yeah, I can see that happening. but password resets? Nope.
It gets... _interesting_ when the FSMO decides to drop off the network. (Or worse, you've done the role seizure on a copy of a DC and it accidentally gets put on the production network and the two FSMOs get into gun fights over who the man is....)
Don't ask how I know this. :)
"Want more fun? Try upgrading IE8 to IE11. (The IE11 download page rejects IE8 visits...)"
... only if you call an aneurysm 'fun'. I managed to use a second machine and a thumb drive to get around that. (at least the stand alone installer is still available...)
Ah, yes. "Zero Tolerance Policy".
When turns a _drawing_ of a firearm into a mandatory suspension or possible expulsion because OMG GUNS WON'T SOMEONE THINK OF THE CHILDREN SOMEONE MIGHT HAVE GOTTEN HURT!!! (Yes, it's happened over here in the US.)
I am *so* glad I'm well and done with formal schooling and don't have kids.
Also, some of us introverts are just functional enough where we do our shopping a week at a time, so we *do* plan our meals a week in advance, so we don't have to get the store on the way home every. single. day and Deal With People.
(I'll spare everyone the side rant of buying foodstuffs in bulk when it's advantageous and using a vaccuum sealer and deep freezer to portion it all out.)
the ISP I used to work for was worse- since we were not on the corporate network, we weren't restricted to what sites we could visit.
You learned within your first few hours to lock the workstation the instant you got up to do anything lest you find your home page set to some really horrible, filthy porn site.
Hilariously enough, I have a friend of mine who has taken it upon himself to pull down a copy of every anime ever produced. *ALL OF IT*.
He's been buying 3 TB drives in pairs for the last month or so.
We went from a Netapp-hosted set of CIFS shares (one head of an active/active HA pair) that were handed out via DFS namespace to a pair of server 2012 R2 boxes (primarily to break the load between home folders and departmental shares for survivability) and in the ~5 months we've had it running, I've had nothing but problems with it. In addition, I've had to build a second pair of servers to get DR capabilities back- With the netapp, we had a second (smaller) filer parked at our DR site and we just used snapmirror to replicate the volumes. Windows server has *nothing* like that, which means I had to implement DFS Replication, which has it's own set of quirks and problems. (for example, it's not really useful for our user's home folders, due to how DFS-R operates and the workload that is imposed.)
The trouble is, I can't go back to the netapp- the cost to reinstate support on it is more than to scrap the entire ~100 TB stack and buy a new one, and then there's the issue of migrating the ~20 TB of data back over in a controlled manner.
Oh yeah- and most of the instability problems were discovered *after* it went into full production, which meant the business saw a handful of outages as we discovered and fixed some of the more egregious ones.
To be honest, all a lock is going to do is keep the drunks [inside|outside] and slow down the professional thief, who's probably already gotten in and made off with the valuables already. (or found that he's broken into the wrong home because there's nothing of useful value)
This. Very Much.
I worked at one place that had several regional offices, all connected to the Home office which was in a separate city some 200 miles (and a 3 hour drive) away.
The bulk of the users had mostly notebooks; Most were very good about keeping files backed up, either putting them on a CD-R or the remote file server when they were in the office.
The boss of one of the regional offices kept his notebook *in* his desk, on a docking station, and came in one morning to find that it had crashed. I got called in, heard the Clunk Of Death, and asked if he had backed up his files to the file server sitting at the Home office. (this remote office had a full T-1 line there, which was pretty decent for 2004)
He told me no, and by the end of the day I had extracted the drive out of the laptop, packed it in as much foam and bubble wrap as I could manage, and was driving it down to the local air freight depot to express ship the door thing off to OnTrack for them to work their $3,000 magic.
And magic it was; there was a single file they could not recover, and it was an OS file. (I had already gone to the local computer store and bought a replacement drive and had rebuilt the machine whilst management was deciding if they wanted to spend the money to recover the drive.)
To drive the point home of 'we know it sucks, but please back up your files to the remote file server', the recovered files were put on said server, and I pulled them across the line to re-populate the machine.
(It was a small-ish company that didn't bother putting a branch file server at each office, although that would have made sense. *shrugs*)
"*Some* versions of Linux. Not including Red Hat, for example."
Technically, it's a support and subscription you are paying for, but that's splitting hairs. (some corporate entities *want* that support assurance and are willing to pay for it, which is why Redhat is where they are.) The software will continue to run after that first year, but if it breaks you are on your own to fix it.
... actually does work ,so long as you take some sensible precautions when doing scheduled maintenance, like patching hosts from VUM or manually updating hosts, etc.
make sure the first host you start with is *not* hosting the vCenter server. I'll get to why in a moment. do whatever needs to be done to it.
Once it's back up, manually vMotion the vCenter box to that host, by itself, at highest priority.
If, for some inexplicable reason you are recovering from something stupid, like the host that held vCenter threw a purple screen of death or some other horrible failure, if you have an idea what host has the vCenter box on it, you can log onto the host directly using the web client or the vCenter client and at least get console access to resurrect the vCenter server.
(been there, done that, worn out the t-shirt.)
I fought with several departments at $company for _years_ to get them to prune stuff down. I finally gave up, primarily because the last network manager bought an absolute shedload of storage which gave us some breathing room. (one of the very few good things the git did during his tenancy)
For DR there's a couple ways to do it; all require a 'warm' or 'hot' site style of DR implementation and replication software installed at either the storage array or hypervisor level. Some applications are able to do site resiliency (Exchange and Active Directory being one I'm personally most familiar with) which is more toward the 'hot site' end of the spectrum.
Done right, it can be done efficiently across a 100 MB link.
For data archival and data corruption issues, I've still been a champion of tape for it's simplicity and durability. (I can 'accidentally' drop an LTO tape off my desk and still ready it afterwards- can't do that with spinning rust, and an SSD in that size is still too expensive for holding backup data.)
I'll echo the sentiments of some of the other comments, however: Companies tend to hang on to every bloody bit of data they make, regardless of if it's still relevant. At some point, you have to let it go.
Raid is for Data protection using inexpensive disks. This PCIe flash device? NOT INEXPENSIVE. (I *might* mirror them if I was super paranoid, but that's just pissing money down the drain)
Now, you want to use them for a classic data cache tiering scenario:
PCIe Flash cards for "hot" tier of storage ( VDI boot storm reduction, super high end database usage, etc.)
Enterprise SSD in raid 1 or 5 for "Warm" tier(Frequently accessed databases and other files)
Multiple Giant Raid 6 Array of spinning rust for "cold" storage.
I've managed to get lucky- the storage arrays we have at $employer are hybrid SSD and spinning rust drives; the SSDs act as the 'hot' and 'warm' tiers, whereas the spinning rust is a raid 6/ triple parity array of slow, large SAS drives. We've not had an I/O problem with them at all in the ~4 years we've had them in use.
That... is a ridiculously overpowered machine.
I'll need five of them.
Both are sort or second hand, though.
1. my boss at $ISP where I used to work related this to me: the site he was at had a power outage, and the generator failed to start up- didn't even crank.. Fortunately, they had massive UPS units, so they had some time to run to the local auto parts shop and buy a new battery for the generator. Turns out one of the overnight NOC staff had a bad battery, and swapped his dude for the one the generator was using.
2. Same company; One of the things I did working for $ISP was travel to their various NOCs doing field work on their routing and switch gear. I was at the Austin NOC, and got the tour by the site manager. I noticed that one of the big DC PDUs had a notch in the power plane and mentioned it to the manager. Apparently, an electrician was working on it using an uninsulated nutdriver and managed to short the power and ground plane (-48v, Many amps) blew the electrician to the wall 30 feet away from the PDU, put the nutdriver *through* the wall, and blew the upstream breaker.
Said electrician survived, but was (understandably) banned from the site.
I have a picture of the notch, too: http://sub-ether.dreamhosters.com/gallery/v/jecook/work_horrors/archive/AUSTIN002.jpg.html
Fortunately(?) I don't have any power horror stories which I witnessed first hand.
"The Register has attempted to contact IBM, locally and at its US headquarters, for 36 hours. The only reply we have received was to refer us to different people inside IBM, who have also not answered questions."
After trying to get support on an obscure IBM branded software product (encryption key management for an LTO tape vault) I suffered the same exact thing; the few people who did respond to me refused to do anything unless I had some obscure contract number (which I didn't have recorded anywhere) nor were they willing to help me divine that information. I managed to grease the wheels via our VAR and get some actual help on the software, but it was a bit aggravating. Unfortunately, the software product is obscure enough that Big Blur is the only vendor for it- no one else seems to want to make a competing product.
In my area, if they ran fiber to the house, it'd probably get trashed by the next installer who decides to use the _line itself to hang his ladder from_. I'd probably get the bill to repair the line, too.
(the 3rd party dude that was sent to do the install for my house actually did that. He was... not very bright, and argued with me that the commercial grade coax connectors I was using would screw up the signal just because the core of the connector was blue instead of black, even though the packaging stated that it was certified for digital cable and cable internet. I decided to let him be and do his own thing, and oddly enough, when I switched my connectors back in the cable modem showed a lot less errors on the line. *shrugs*)
Mumps (now called Cache) is a horror show and a canonical example of how NOT to design a programming language. (It makes the Obfuscated C contest look verbose in comparison)
(Google "The Daily WTF mumps" for several articles about it)
Dust off, Nuke it from Orbit, it's the only way to be sure.
Windows 98 Second Edition actually had proper USB 1.0 support.
There was an 'OEM only' release of windows 95 (known as Windows 95 C) that had rather limited USB support.
I still consider Windows ME the b@$tard offspring of windows 98 SE and Windows 2000- the worst of 98 SE, with the pretty pretty skin of 2000.
My last manager was like that, (the 'too fast and too furious') only kind of like Godzilla:
He came, He f*@ked up all our stuff, and then left just before it all blew up in his face, leaving us to clean up the mess and make some sense out of what he did.
Thankfully, his replacement is *much* better and has at least half a clue and a willingness to listen to his staff when they say 'hey, we tried that four years ago and it sucked then and unlikely improved in the interim'.
Beer, because it's friday, and I'm going to need one to get rid of the PTSD flashbacks from Godzilla manager.
... I'll be happy if M$ created Service Pack 2 for Windows 7 and realeased it, TBH. I should not have to kill half a day and 4-5 reboot cycles installing updates after a fresh install of W7 SP1.
(You remember those, right? MS rolled up all the various patches, bugfixes, and updates into a single cohesive package that you downloaded, installed, and rebooted once to update a fresh from install media to a reasonably recent point in time to keep from going through the update process 4-8 times? Or even better, could merge with the installation media to make life easier? I miss those, too.)
Beer, because it's friday.
Biting the hand that feeds IT © 1998–2017