The Grumpy IT Admin

Some more thoughts on SSDs you may wish to take on board. To maximise the life of your drive you need to minimise the amount of deletes that go on, try to set up your PC so that you don't have too much data changes going on. Easy targets are Hiberfil.sys and Pagefile.sys. Hiberfil.sys is basically a replica of what is in your RAM when you go in to hibernation mode. Now I’m not sure about you but I never use this. Under XP this was a simple turn of under Power Configuration in control panel. On Windows 7 you need to get a cmd box up with Admin rights (right click Command Prompt and select Run as Admin) and type powercfg -H OFF, hit enter and your Hiberfil.sys will disappear never to be seen again. Moving the Pagefile is little more involved. You can remove completely if you wish but don’t shout at me when Windows falls over or glitches. Not having a page file will cause problems and not having a page file of the right size in the right place can also cause problems. If you don’

I have just gone through that buttock clenching experience of a firmware upgrade on my SSD drive. This should be a simple and easy task, I have done it many times on ordinary disks from within a Windows session, indeed firmware upgrades for all things come thick and fast in the land of HP servers, you download the ISO run the update program and all sorts of firmware updates happen it gets on with it and then you reboot and all is well again. BIOS updates for motherboards even happen from within Windows these days, unfortunately it seems that Micron, care of it's Crucial arm, aren't up with the project. To update a C300 SSD you need to download the ISO, burn it to CD then change your SATA settings in your BIOS to IDE only, AHCI will break it. you then need to make sure your SSD is in channel 0-3 or the update program won’t see it. Then you boot off the CD and run through the program, typing yes at the appropriate moment. Did I mention you need to back up your data first? The

On price per GB certainly not but then what do you use your storage for? In the enterprise they always talk of tiered storage. You put the disk intensive stuff on the fastest drive you can afford, then the rest goes on something that is reasonably quick but cheap to buy in bulk, then you go for something that is so cheap per GB but slow as wading through porridge where you only put the really old stuff, then there is archive to tape where you are talking minutes/hours or even days to get back to the user. In old days you had SCSI, IDE and tape, the disk were arrange in to RAID arrays, these stated simply with stripe sets of a few disks and mirrored pairs of disks, then you got redundant disks with raid 5 just in case something failed. Typically a database box would have system on a mirrored pair, Logs on mirrored pair and DB on a raid 5 of 3 or more disks depending on the size of the beast.  Then along came SANs but these were just glorified storage shelves with the disk arranged

I had the joy of finishing off an LTO 5 library install today. It is a bit of beast with 96 tape slots available and 4 drive bays. First problem I had was that the robot arm was stuck after powering it on, so power off, pull a couple drives out and see if anything is blocking it, nothing obvious, so it gave it a bit of twiddle and prod, powered back on and everything is splendid. We have the 4 drives but in a typical HP mental licensing move we are only licensed for 48 tape slots. So they charge you to buy a load of hardware then make you pay a licence fee to use it all. Imagine if you bought a Truck from but you only were allowed to use the first half of the trailer unless you paid an additional license fee, would you   think you had got a good deal? How about if boughtV12 Ferrari but your initial licence only allowed you to use it as V6, but you could pay Ferrari an extra wodge for the other cylinders if you needed them later. A mad way to do business. Anyway, the tape drives use

Our DR EqualLogic has relocated to our remote site and is up and running. there is a whole bunch of networking that need to be done, but now the we have  Dell Sumo unit on the end of a 1Gbps link and our production units at the other end with replication going between the two. Somehow between BT and Virgin Media we have managed to get jumbo frames working between the two sites and with 3ms delay each way on the 1Gig link we actually saw it hitting the full 1Gbps speed, which is not what we would have expected, anything in the 5-800 would have been desirable, but with the 6ms round trip we were expecting to see around 250 or less certainly our testing with WANem indicated a much slower replication, though that was with 5 EqualLogic units (1 SATA and 4 SAS) replicating around 20 volumes. Replicating just the one or two volumes saw a much lower transfer rate. We will have to see how it goes over the next few weeks, especially once the VMS boys start replicating their EVA stuff. The i

Not much going on at the moment. The errant EQ is up and running OK, and no suppliers have managed to upset me, apart from the usual external developer. They seem incapable of developing in any meaningful way, they develop on their own systems, they move the new bits to our QA systems, then they go live and everything stops working. Do they do any load testing, no. Do they have any meaningful way of testing their LOB in any way for incompatibilities, no. Do they have any way of taking advantage of a 64 bit OS, no, they run .NET in 32 bit mode. Do they have the ability to exploit modern multi-core processors that they rely on in the physical servers they insist we use, no it is all single threaded applications all the way. Do I laugh, no because it is my fault it doesn't work. It is my hardware, my OS, my storage and my network, nothing to do with the supremely efficient and well manged code, even after they update it, yes they log on to our servers and update our live systems,

Had my car insurance renewal to day, I drive a Nissan Primera, 2L auto, I am 46, and live in a rural area, they want £1355 for TPFT, yes that's right there's no missing decimal point in there. Had look at the comparison web sites and I can get fully comp for £280, even the same company would only carge me £290 on the websites. The best bit was the covering letter that stated they "strive to keep prices down" obviously they need some lessons in striving as they are not very good at it.

Graphics cards with multiple outputs are a bit of a bug bear at the moment. Now these beasties are almost quite clever, as when you boot your PC up they will look for a connected screen and display on that. Ah but what about if no screen is connected or in my case the screen is turned off when you boot the PC? My logic dictates that it would always default to one of the outputs, perhaps output 0 or some such, but no in the case of my ATI 4870 it seems to randomly select an output. This is very annoying as I have a power save 8 way adapter that powers off all the sockets and thus all the peripherals when the PC is shutdown. When I switch the pc on the monitor does not automatically come on and thus never sends a signal to the PC and I end up under the desk swapping the lead over to the other output. All very unnecessarily irritating. Another one is the Xbox, when you get one of these you want to play so you create a profile and get on with gaming, then you think I just link this pro

Well that was a day and half Started at 05:30 this morning, first early shift in the rotation and I never sleep well, awake at 4am with usual alarm clock not going off panic. Dell engineer due in at 8am, did a quick firmware upgrade on our out of sorts EqualLogic sumo. A triple bypass is in the pipeline – disk, controller and SATA card. Engineer turns up at 11am, not actually a Dell man but a sub-contractor, nice enough but filling me with confidence. Disk replacement goes well, well I could have done that, controller not so perfect but get ther in the end after a quick call to Dell EU central in India,  who then talk us through the SATA card replacement. This is where it all went a bit Pete Tong. The support guy was telling us to pull the left hand card the unit was saying right and omg….. It’s not fallen over.  However when we enabled the SATA card the brown stuff hits the fan. Suddenly we are looking at 6 failed disks. And an array that is off line and all our DEV and QA systems

Well I managed to fix the remote backup server and all has run ok overnight. In the end we just added in an extra iSCSI drive copied over the BE catalogues which took 3 goes and 3 reboots as the drive kept failing. Still they now have some working backups though they seem to be backing up server that we don't have back up licenses for so that will be the next thing to sort out along with setting up the jobs properly. Back on our own Backup jobs I came in to a typical piece of BE stupidity this morning, We have 4 drives in our tape library and BE had decided to put duplicate tapes in to three drives leaving only one drive available for the originals. so two jobs were just sitting there doing nothing waiting for the duplication task to complete, no wonder our backs seem to take for ever. Why the job scheduler can't check that all media and drives that are required for the job are available before loading up tapes is beyond me, this is more of a problem for backup duration than

Couple of Backup Exec quickies, one tip and one question Tip If you run the live update and the download stops at 3000KB then the problem is possibly that you are going through a proxy, there's nothing wrong with your proxy, it is Live update which for some reason will get to 3000KB then suddenly decide to bypas the proxy and go direct. Why? nobody knows and Symantec seem deny it is happening. There is a fix, you go in to control panel, in to the Live Update settings. On the HTTP tab configure the proxy manually rather than letting LU work it out for itself, which it is incapable of doing. This maybe a issue with IE 8, as I had IE9 beta on my PC and it was fine, but when I rolled back to IE8 I was suddenly afflicted by the issue, also a server that had IE7 never had the problem but sometime after IE8 was installed it started going wrong. This might be a red herring as Iit is just observations. Question Why when you move the BE DB, or try to upgrade it or do pretty much anyth

Why is it that people don't seem to be able to do things right first time? One of the most important aspects of the  System admin role is to make sure you have backups, and good backups at that, and if at all possible copies of your backups held many miles away, on different types of media.... One of our sites has had some problems with their backups, one guy got halfway through fixing it before disappearing off on holiday leaving no backups over the holidays and no end of year backups. Wouldn't mind being a fly on the wall when he has responsibilty explained to him on his return. It was suggested that the remaining tech bloke on site should probably be looking to rebuild the server starting again from scratch, this drew the puzzling response of I don't know how to rebuild a server. That bang was my jaw hitting the deck. Anyways it now seems to fall to me to fix the problem of a server that has issues with one of its online disks and a Backup Exec (why did Symantec buy a

What is it with new kit that makes it not work properly? I often find once you make it through the first month you don’t tend to get any failures for years. Our most recent EqualLogic is a case in point. This is a P65000 or Sumo unit with a whole bunch of SATA drive in it. First off it was ok but a drive went after a week, then another drive went, then the same drive went again, prefaced by a large number of controller can’t talk to drive, yes it can, no it can’t, yes it can logs. Then the replacement went the same way. Now to my mind this indicates that something fundamental is wrong. First thing I was asked to do by the dell man was to failover the controllers, now there is no obvious way of doing this, no big button that you can click on, so a quick Google and it seems that the way to do this is to restart the unit. Now if you are like me Restart means shutdown then start up again, but in the EQ world it means that you shut down the controller, failover to the other controller th

I seem to have a mental aberration every month or so, something to do with full moons I think, when I do something rather rash, often preceded by doing too many things at once and time constraints. During the last one I decided to set the wrong drive offline. This caused a couple of servers to disappear, no VMware is really quite resilient but losing a drive seemed to cause it a few problems. I quickly realised my error and set the drive online again, however, the two ESXi host servers that had running guest servers on the offline volume had locked up and were unresponsive, though other guest servers on the two host servers will still up and running though the VC was showing the ESXi servers as disconnected and the other servers as unavailable. I started shutting down the servers using RDP to connect with the intention of forcing a reboot of the ESXi servers. I cleared one off and restarted it in the meantime the second one had managed to recover its composure. All the servers wer