Have you tried turning it off and on again?

IT Crowd - Roy

Yosemite has been playing nicely on my MacBook Air so I decided it was time to upgrade the iMac.

After doing all the usual safety procedures and testing the backups it was time to upgrade. The iMac has had a few OS X upgrades since I bought it – it came with 10.6 Snow Leopard – and I did consider doing a clean install. However the “normal” upgrade route has had good reports so I opted for that.

The upgrade proceeded smoothly to begin with and I left it to get on with it, however when I returned some time later it seemed to have stuck on the last install screen at 50% complete. I left it for about half an hour and nothing had changed so I did what any highly trained Mac technician would do.

I powered it off and on again.

That did the trick and the iMac booted happily. I had been pretty sure it had got stuck at the reboot stage of the process so turning it off and on again forced the reboot it should have done by itself. Just to be on the safe side I then restarted it up in Safe (or Single User) Mode (Cmd + S) and ran /sbin/fsck -fy. Next, I booted from the Recovery Partition (Cmd + R) and used Disk Utility to repair permissions on the iMac’s startup disk.

Since then it has been fine and I’m now in the process of reorganising my data files to take advantage of iCloud Drive.

Geocaching once more

In which I get my geocaching mojo back.

Two months after my retina detached itself, curtailing my geocaching activities, I finally made it out onto the footpaths of Bucks.

Obviously I can see where I’m going but I was sceptical about my ability to spot a nano in an ICT, so I decided that I’d go for slightly larger containers on my first outing. I’d been following with interest the fate of some Chiltern Hundred caches which I had previously DNFd. Fortunately, my fellow cacher (and geo-blogger), Washknight, had started to work his way round the series and was actively undertaking maintenance when necessary. So I figured if a blind man could manage it, a one-and-three-quarter-sighted bloke like myself should have no problems. Actually (and of course you’ll know this if you read his blog) Washknight is actually Paul, his wife/partner-in-crime and his son, but you get my drift. :)

So, having parked up somewhere familiar once more, Skye and I set off along a footpath towards CH002 – Chiltern Hundred, bridge. This one was a bit out on its own for us but a necessary find for completeness. This turned out to be the smallest cache of the trip and it did involve Ivy! After that we retraced our route towards the car, picking up CH005 – Chiltern Hundred, Chesham back and CH003 – Chiltern Hundred, Chesham heights. Stopping for coffee and a biscuit by the car, we then set off to find CH008 – Chiltern Hundred, bridleway. Last time we were here, the bridleway was a quagmire of gloopy mud and Skye got completely covered in mud. It was such heavy going that I ended up carrying her. And of course we couldn’t find the cache.  Since then the path has been resurfaced and we found it to be mud free. Until we got to within 50 yards of GZ. There must be something strange going on from a geological point of view because that section of path was very muddy. Significant churning up by passing horses didn’t help the situation. This time I decided to pick Skye up before she got muddy and I’m sure she appreciated the lift. When we got to GZ I could see that there was only one place that the cache could be and it was. Very pleased to get a quick find. As we returned back the way we came, we made a detour into a small wood (owned and looked after by The Woodland Trust) which bypassed the muddy section of path. Result!

The next two caches I wanted to find were up near the village of Botley, so we went back to the car and drove there, parking in the carpark of the Hen & Chicken pub. From there it was a short walk to CH012 – Chiltern Hundred, Lee farm although spotting the footpath sign pointing into what looked like a private farm yard took me a few minutes. Finding the cache took a further few minutes but having the knowledge that Washknight had been there and maintained the cache just a couple of days before gave me the confidence to a) search in an exposed area and b) ignore the many muggle vehicles which drove past me. Of course nobody paid me the slightest attention! After that it was on to the last one of the day, CH013 – Chiltern Hundred, Codmore view. The funny thing about this one was that Washknight had found 2 containers on his visit. He performed maintenance on the original container and removed the other, temporary, one. I was therefore surprised to spot the container in a place which didn’t fit the hint and was a few metres off. A look at the log showed me that this was yet another “throw-down”, albeit a nice container and not just a film pot. Shortly afterwards I located the correct, original, container with Washknight’s new log and laminated card with the bonus code! So I signed that log and removed the other container. I wonder if there are any more?

I was pleased with the day. Skye had a really nice long walk and I got my caching eye back in. I filled in a matrix day and my six finds brought my total Chiltern Hundred finds up to 99, excluding the bonus. I really just need to find one more of my DNFs and that will be sufficient, I think.

Lastly, a big thank you to team Washknight for their cache maintenance which certainly helped to make those DNFs more “findable”.

Seagate HDD fixed

The adaptor I ordered arrived after a couple of days, so I set about fixing the problem. A spot of the good ole google and I found several web articles which provided me with the information I needed to attempt the repair. I’ve reproduced the steps below in case you are interested in seeing how I did it. Plus, in the unlikely event that you encounter the same problem as I did, you won’t have to search for the fix elsewhere.

There are two common firmware errors on the seagate 7200.11, these are:

0 LBA error – when you computer recognises the HDD but reports a size of 0 bytes

BSY error – when the HDD enters a BuSY state and is not recognised by your computer.

My drive suffered from the latter and the following instructions address that error.

Firstly, a warning. It is very easy to kill a hard disk so proceed with caution. This procedure worked for me but it may not work for you and I make no promises or guarantees that it will. I am NOT responsible for any damage or data loss. In other words, you’re on your own!

First job (if you haven’t done so already) is to extract the hard disk (HDD) from its external case and power supply. Then, use a TORX screwdriver to remove the screws fixing the HDD PCB to the chassis.

HDD exposed

With the PCB removed, you can see the contacts for the HDD motor (middle) and the contacts which control the HDD’s operation (lower right).

HDD contacts

You need to isolate the controller contacts. Place a suitably sized piece of card over the contacts and replace the PCB. Fix it in place by replacing the top 3 TORX screws and tighten them. Replace the other 3 but leave them loose.

HDD with card isolating drive contacts

You need to provide power to the HDD. In my case, I removed the second PCB (which controls power and data interface to the computer etc.) from the external case and connected it to the HDD. Luckily the HDD’s serial connectors were still accessible. In this photo you can see the HDD PCB replaced with the card isolating the controller contacts, on the right of the photo is the second PCB with the A/C power attached.

It is worth turing the power on briefly to check that it is supplying power to the HDD and that it spins up. Disconnect the power for now.

HDD with power PCB attached

The USB/RS232 adaptor has 5 pins. The outer two supply power and are not used in this setup. The adaptor needs to be tested without the drive attached. Without any cables, insert in into a free USB port on your computer and run up your terminal program. For this procedure I was using a DELL laptop running Vista. Yes, I know, Me using Windows! There was no way I was going to risk frying my MacBook with some untried procedure and this old DELL was fair game!

I downloaded hyperterminal and configured the port which Windows assigned (in my case COMM3) as follows:

  • Baud 38400
  • Data Bits 8
  • Stop Bits 1
  • Parity none
  • Flow Control none

Test the adaptor as follows. Type a few characters. If you see them displayed on your screen then your terminal program is in local echo mode and you need to switch that off. Next create a loopback connection. Simply connect the Tx and Rx pins. Now when you type you should see the characters displayed as you type. If all is good then you can proceed to fix the drive.

RS232 to USB adaptor

Now connect the HDD’s Tx, Rx and ground pins to the RS232 adaptor. To get the cables into the small space on the HDD I had to remove the connector’s outer insulation. These were just plastic mouldings which came off easily. Once the connections were made I used small pieces of insulating tape to keep them from touching each other or moving about. Ground goes to Ground, whilst the HDD’s Tx is connected to the adaptor’s Rx and the HDD’s Rx to the adaptor’s Tx.

This photo shows the HDD’s Tx and Rx pins with the cables plugged in before taping them up. The pins from right to left are Tx, Rx, Ground, unused (as far as we are concerned).

HDD Tx and Rx connections

Now for the tricky part.

Power up the HDD. Your USB will supply power to the adaptor. The commands you need to type are in red bold. Please note they are case sensitive.

After a few seconds Press Ctrl + Z

You should see this prompt on your terminal screen:

F3 T>

If not, swap the Tx and Rx cables and try again.

Now access Level 2:

F3 T>/2 [press Enter]

F3 2>

Wait about 20 seconds then spin down the HDD motor:

F3 2>Z [Enter]

Spin Down Complete

Elapsed Time 0.146 msecs

F3 2>

If you instead see a message something like this:

LED: 0000CE FAddr: 00280D4D,

you entered the commands too quickly. Power off and on, wait 20 seconds and start again.

Next, carefully remove the piece of card you placed between the PCB and the HDD contacts and tighten the remaining TORX screws. Spin up the drive motor:

F3 2>U [Enter]

Spin Up Complete

Elapsed Time 7.087 msecs

F3 2>

Now access Level 1:

F3 2>/1 [Enter]

Do a S.M.A.R.T erase:

F3 1>N1 [Enter]

When the prompt comes back, turn off the power to the HDD, wait a few seconds then turn it on again. Wait about 20 seconds then do a partition regeneration. If you don’t see the prompt do:

Ctrl + Z

F3 T>

Enter the partition command carefully:

F3 T>m0,2,2,0,0,0,0,22 [Enter]

If all goes well you should eventually see a message something like mine, below:

Screenshot showing successful completion

Do not turn the drive off until you see this message. Once you have seen it you can power everything off and reassemble the drive, etc.

And that’s it. If you are lucky your drive should be readable and you should be able to access the original data. At this point you have several options, e.g. copy the data somewhere else or at least back it up.

What happened with my drive? Well one of the partitions had missing data while the other seemed intact. As my data wasn’t unique or remotely critical, I decided not to bother with data recovery and instead plugged it into my Mac and used Disk Utility to repartition it as a new drive. It seems to be behaving itself so I’m using it as a network store and a redundant disk where I can copy non-critical files.

Seagate drive problem

This evening I opened up the failed hard drive with the intention of destroying the physical disk before taking it to the dump for recycling. I discovered that despite the Iomega branding on the case, the actual hardware was a Seagate drive.

When I realised the disk was a Barracuda 7200.11 I thought, “Hang on, I have heard about these”. I remembered drsolly writing about these on his blog so I read that again and then did a bit of research of my own. The problem with these drives is that if you start them up when the internal event log is at line 320 (or something like that) the disk dies, even though all the data is still there on. Anyway, there are a couple of fixes which may (or may not) work in my case so I’m going to have a play with trying to fix it. I need to connect to the disk from a serial port so I’ve ordered an adapter.

I’ll report back when it arrives!

This is why we back up our files

I just had an external hard drive fail. This is an old disk, in fact it’s the oldest one on my LAN. It’s an Iomega 500MB (a huge capacity when I bought it!) which was connected to my router as a network drive. This morning it crashed while I was copying a file from it. Later, when I had time to investigate further, it failed in a major way. I tried some “first aid” using Disk Utility but I already knew it was a dead-un. I managed to copy some important photos off it before it ground to a complete halt and that was it!

Luckily there wasn’t anything important on it, mainly duplicate backups and copies of OS updates, patches, etc., to save me downloading them again if needed. Even the photos were just just local copies of ones that I am also storing online and elsewhere. I say “luckily” but it was more by design than luck.

So although it was annoying, I shan’t spend any serious time on recovery operations. I wouldn’t feel I could trust the disk afterwards anyway.

What this event has done though is reminded me that it is not a question of whether a hard disk will fail but when.

So not only do you need to backup and backup often, keep those backups on more than one hard drive and keep a copy somewhere on a cloud-based storage service. Check that you can actually restore from said backup. Keep an eye on the age of your main backup drives and replace them when they get a bit long in the tooth.


When Yosemite first dropped I wasn’t in the mood to be one of the “early adopters”. Normally with a new OS I’d be straight in there but I put off updating from Mavericks and I think I did the right thing. Apparently there were lots of minor problems reported on the discussion forums, not least a problem with dropped WiFi connections. Anyway, I decided I’d wait for a point release and then of course I had my eye problem…

Now that Apple have released a further point release (OS X 10.10.2) which purported amongst other things to fix the WiFi issue I thought I’d update my MacBook Air for starters. Opinion is divided between those who simply update from whatever their existing OS was and those who insist that a clean install followed by a restore of their backed-up data is the only surefire way to a reliable system. I thought I would risk the former and use my backup plan to do a clean install and restore if necessary.

I followed my usual backup method – which is really about ensuring I have a good backup in case of a SNAFU – prior to starting the update. In other words:

  1. Backup to external disk using Time Machine
  2. Create a bootable clone of the entire Macintosh HD partition on an external disk (I use Superduper!).
  3. Test the clone by booting from it and testing the basic functions of OS X.
  4. Verifying the Macintosh HD disk.

No. 4 gave me a couple of disk errors. Nothing serious. Nevertheless I booted the Mac in safe mode (that’s booting it while holding down the CMD and S keys) and running fsck to check and repair the disk:

/sbin/fsck – fy

It found some minor file allocation errors and then returned the message “Macintosh HD appears to be OK” (which is always a good sign). I rebooted it as normal and downloaded the Yosemite installer from the Apple App store.

It pays to have a Plan B, which in my case would be to restore everything to its last Mavericks configuration – I have a backup for that! – but it’s very useful to have a stand-alone copy of the OS X installer on a USB drive. If you have that, you can boot from it and run diagnostics or repairs from there and/or install the OS from scratch if need be. A useful spin-off is that if you have several Macs to update to the new OS, having a copy of the installer on a portable disk saves having to download the large installer image onto each Mac.

Starting with Mavericks, Apple have made it easy to create a bootable copy of the installer. There are other ways of doing it but I think using Apple’s “createinstallmedia” utility is fine. Here’s the method:

  1. When you download the OS X Installer app, it places it in the Applications folder. This method assumes you’ve left it there. (Don’t forget, if you run the installer and update the Mac, the installer app gets deleted from the Applications folder afterwards, so you need to create the USB copy first).
  2. Using Disk Utility, format an 8GB USB drive and give it the label “Untitled”. Leave it connected to a USB port on your Mac.
  3. Open a Terminal window and enter this command:

    sudo /Applications/Install\ OS\ X\ Yosemite.app/Contents/Resources/createinstallmedia –volume /Volumes/Untitled –applicationpath /Applications/Install\ OS\ X\ Yosemite.app –nointeraction

  4. Let it run (takes a while!) and you will be left with a bootable USB disk containing the installer.
  5. To start your Mac from the USB drive, start the Mac whilst holding the Option (alt) key down. Once the Mac starts it will display the list of available bootable drives, select the USB drive from that list.

So, having tested my backups, downloaded the OS X update and created my USB installer, I was ready to undertake the update.

The Yosemite update was very straightforward (as Apple intended) and my Mac was soon up and running. I’ve left iCloud Drive off for the moment until I’m happy that everything is working as it should.

Cache maintenance

I’m not sure how successful finding caches will be at the moment, so instead I’ve done some maintenance on a cache of mine which I had to disable back in November.

My cache GC56V75 – Buncefield’s Big Bang was originally placed in June 14  to celebrate the reopening of Cherry Trees Lane which had been closed for 8½ years. Fast forward to November and the contractors redeveloping the storage tanks either side of the lane decided to temporarily close the lane.

Bit annoyed.

Anyway, I “temped” (temporarily disabled) the cache, intending to relocate it. Unfortunately I suffered my detached retina before I had time to hide the replacement so that never got done. Until now! Yesterday I found a new location and hid a new container. When I tried to edit the coords on the cache page I found the distance from the original was too great for the system to accept the change automatically. Instead I had to email our local reviewer to ask hime to change the coords for me. This evening he emailed me back to say it was done (many thanks to Red Duster for turning it round so quickly, ) so I have now been able to enable the cache once more.

On the subject of maintenance, I was thinking I really should go and check on another of my caches – GC2XEJE Short Walk, find the Nano – which hadn’t been found since last September, when I was pleasantly surprised to get a “found” email. Someone had logged a find. Yey!

That saves me having to go and check that one until it’s convenient for me. ;)


Get every new post delivered to your Inbox.

Join 249 other followers

%d bloggers like this: