deKay’s Blog

I went to use my 360 this evening, and as soon as it powered on and logged in, it froze. Restarted it, and the same happened again. Zomoniac off of ugvm reported similar problems yesterday.

After some faffing around, I tried disconnecting my network cable and starting the 360 up again - and this worked. Plug in the cable, connect to Live, and it freezes. Naturally, and following the problems with Live all week, I assumed that it was Microsoft’s servers up the spout.

Then a post on rllmuk linked here. Turns out than an incomplete download of the new Pro Evolution Soccer game is to blame. And what was I part way through downloading when I switched my 360 off last night? Exactly.

So I disconnected the 360 from the network once more, started it up, deleted the incomplete demo download, and reconnected to Xbox Live. The result? It works once more! Phew!

For the last couple of days, Outlook has been throwing up the following error:

Task 'Microsoft Exchange Server' reported error
(0x8004010F): 'The operation failed. An object could not be found.'

Googling it wasn’t much help, as there seemed to be more than 20 reasons for the fault. Emails were syncing OK, and everything else appeared to be normal. However, checking the Exchange (2003) server, and looking at the Offline Address List settings (alluded to in the majority of the search results), I found another error message. When trying to rebuild the offline address book, this came up:

There are no bindings.
Facility: Win32
ID no: c00706b6
Exchange System Manager

This gave me a bit more to go on, and thanks to this page, I eventually found this solution on Microsoft’s website.

Basically, I deleted the old offline address list, created a new one, pointed it at the Default Global Address List, then waited a short while for the servers to catch on to the change. I could then set the new offline address list as the default, and then rebuild it.

Then it was a case of waiting for the next list sync (I triggered it by changing the update interval for the offline list to a custom one forcing an update in the next 15 minutes) and then quitting Outlook on my machine, reopening it, and then downloading the address book from Tools > Send/Receive > Download Address Book…

Phew!

Please note: this fix requires doing things to your iPod that you wouldn’t normally do. Even though I’m certain they’re safe, I can’t be held responsible if you knacker it following these instructions!

How frustrating. After almost a year’s loyal service, I went to charge up my 3G iPod nano before a trip, only to find it wasn’t working. It would turn on, but after about 5 seconds of showing the menu, it would reset itself. Over and over again.

I blamed the 1.1.3 firmware, which I installed a few days ago. I hadn’t used my iPod since then, so probably wouldn’t have noticed the problem. Oh well, I thought, a factory reset should sort it.

But how to do a factory reset? It would appear the only way of doing so, is from within iTunes. Sadly, since my iPod didn’t stay running long enough for iTunes to even realise it was plugged in, that was no answer. Unlike many devices, the nano doesn’t have a little hole with a paper-clip-able reset button in it, so that was no option either. All I could find on the internet were how to soft-rest it (hold down Menu and Select for a few seconds), which didn’t help (it could soft-reset all by itself anyway!), and how to stop it reset looping (let the battery drain).

So having left it looping for a few days, the battery was finally flat. I plugged it in, turned it on, and… it reset itself again.

Then I was pointed to this page. It shows how to enter the nano’s diagnostics menu! You reset the iPod, then immediately hold down Reverse and Select, keeping them held down until you get a screen like this:

As interesting as it is (and different to the one described in the link above), none of the options listed actually helped in this instance. There isn’t a factory reset here.

Then I had a thought - if Menu and Select does something, and Reverse and Select does something else, what about the other combinations? I tried Forward and Select, but that did nothing. However - Play/Pause and Select (held down immediately following a reset) did this:

Which not only didn’t reset after 5 seconds, but also allowed iTunes to sync again, and let me perform a factory reset! It seems that particular button combination forces the iPod into GUI-less disk mode. Excellent!

I set about resyncing all my music and podcasts and stuff, and all seemed well. I realised that, by default, some things are not automatically synced - games, videos, contacts and calendars in particular. So I added those to the sync lists and they copied over to the iPod.

Disaster! The resetting recurred!

I repeated the whole process. Wiped the iPod, resynced my music, then podcasts, then contacts, then calendars… and reset. It was the calendar syncing that screws it up! I put the iPod back in forced disk mode, removed the calendars from the sync, and resynced. It worked! Put the calendars back, it resets!

I have no idea why this is happening, although my upgrade of SyncMate recently did cause some event duplication in my calendar that required a Time Machine restore, so perhaps that introduced something. It doesn’t really matter as dates and stuff aren’t necessary on my iPod, so for the time being I’ve just disabled them.

Anyway, hopefully this might help others who are tearing their hair out with iPod nano resets!

Fo some reason, my XP Pro install in Parallels on my iMac lost internet access a few weeks ago. I wasn’t bothered as I hardly use it for anything. Today, however, I needed it as part of the install of Amiga Forever 2008, which only really works in Windows due to there not being a decent Intel port of UAE for the Mac. But I digress.

Having tried all the settings in Parallels, resinstalling the Parallels Tools in Windows, rebooting Windows, and repairing connections, I was still getting nothing. Supposedly an update to Leopard (10.5.3, I believe) can cause the Mac OS firewall to block DHCP requests from Parallels hosted OSes, but the firewall was turned off and XP could get an IP easily enough.

There has also been a recent XP update that can cause HTTP access to seemingly die, particularly if you have ZoneAlarm installed. However, I haven’t installed this update, and don’t have ZoneAlarm on the XP install. Besides, I’ve had no internet access on XP for a while longer than the update has been around.

Eventually, I found an almost-fix. Almost in that Internet Explorer on XP now works, and I can ping addresses, but for some reason Firefox still can’t see the internet. Anyway, it’ll do for now.

This post here provided the answer, which, if you can’t be bothered clicking, basically requires you to type this into your Mac’s terminal:

sudo killall -HUP pvsnatd

I’ve had a HP laptop in several times in the last week, suffering from severe audio stuttering - in Media Player, VLC, when playing DVDs, and even just playing the Windows start-up jingle. It’s almost like when I had a nasty 16-bit ISA soundcard in my ancient PC and tried to play Half-Life.

We tried all sorts - scanning for viruses, defragging, reinstalling drivers, reducing the hardware acceleration, adding more memory, changing the volume… everything. But it still kept doing it.

I did notice that in Task Manager, whilst playing MP3s in Media Player, svchost was spiking at around 80% CPU usage, and Media Player itself flicked between 30-odd to 80. Playing the MP3s in other media players showed a similar thing. It was all very strange.

Thankfully, the ever amazing internet and some carefully chosen search terms married up to provide a solution. It would appear that some HP laptops of similar age to this one, as well as other machines (notably Sony Viaos) are prone to suffering from audio stutter. The reason? A mis-detected hard drive mode!  It seems that XP is seeing them as PIO rather than DMA drives. You can change the setting manually, but after a time, XP forgets and the cracked sound returns.

This forum topic explains it a bit more, but, more importantly, is what led me to this page - complete with a VBS script to fix the problem once and for all!

I was asked to fix a problem with an IMCE/TinyMCE/Drupal 6 for someone today. Basically, the upload button wasn’t appearing in the popup box when you click the image icon on the TinyMCE toolbar.  This has happened to me before, on Drupal 5.x, and was because I hadn’t given the relevent users permission to use IMCE, or configured IMCE to be used with TinyMCE.

In the new version of IMCE, the latter option is no longer available, and everything else was set up correctly. After much testing and searching, it turned out that the answer was to simply read the readme file that comes with the IMCE files. It points to this page here, with some PHP script to add to the current theme’s template.php file. And now it works!

Now, if anyone can explain why this is now necessary when it wasn’t previously, I’d love to know.

I came across this strange thing today. One of the domain controllers wasn’t picking up new group policies. Basically, there were fewer folders in the policies folder in SYSVOL on that DC than in the others.  Looking at the event log, the following was thrown up:

Event Type:    Error
Event Source:    NtFrs
Event Category:    None
Event ID:    13568
Date:        06/06/2008
Time:        12:17:46
User:        N/A
Computer:    GWEN
Description:
The File Replication Service has detected that the replica set “DOMAIN SYSTEM VOLUME (SYSVOL SHARE)” is in JRNL_WRAP_ERROR.

Yes, the server is called Gwen. This is A Story for Another Time.

Restarting the File Replication Service did nothing. Neither did rebooting the server, or disabling file replication and then re-enabling it. Thankfully, though, there is a pretty simple fix.

It turns out that the journal used for the replication service had become corrupt, preventing the replication from happening. What you need to do is force the journal to reinitialise so that it rebuilds the folder again from scratch - not a problem if your other DCs have up-to-date SYSVOL folders. This reinitialisation on Windows Server 2000 is automatic if “jrnl_wrap_error” happens and Server 2000 SP2 is installed, but (curiously) not if SP3 is installed.

You can, however, force a SP3 machine to reinitialise (or, as Microsoft put it, “automatic non-authoritative restore”) the share by modifying the following registry key on the server:

In HKLM\System\CurrentControlSet\Services\Ntfrs\Parameters, change the “Enable journal wrap automatic restore” key to “1″. If it doesn’t exist (it didn’t for me) add it as a DWORD key, value 1.

Then, restart the File Replication Service from Services, and wait! The folder is backed up (the event viewer tells you where), deleted, and then re-synced with other servers.  Hurrah!

Please note that modifying your registry is dangerous and may cause your feet to explode.  Or something.  Read Microsoft’s help page on this issue, which explains a bit more and provides another possible solution.

After upgrading to v1.2 (v1.2.0.325, in fact) of SyncMate, my Vario II (Windows Mobile 5 phone) decided it no longer wanted to talk to my Mac. Either of my Macs, in fact. I kept getting the message “Can’t start driver, error 2″.

The “fix” suggested (deleting some files on the Vario II, reinstalling the driver, etc. - see thread here) then prevented the Vario II from talking to ActiveSync on my PC at work too.

An uninstall of v1.2 on my Macs, a hard reset of my Vario II, a reboot of the Macs followed by a reinstall of v1.1.0.316 of SyncMate (and another reboot) and finally reconnecting my phone up to install the driver on it (and then rebooting the phone again) finally sorted it.

Being a netrock manager (like a network manager, only more metal), I often need to deploy software across hundreds of PCs. Helpful companies provide “msi” files which I can just plug into a Group Policy on Active Directory, but less helpful companies insist on using exe installers, using InstallShield or WISE or whatever.

These companies need to be punched in their actual faces.

Luckily, some of these installers actually include msi files within the install exe, but actually getting to them is a bit hit and miss - WinRAR will sometimes extract exe files if it realises they’re really just archives with an exe bit tacked on, but sadly this isn’t always the case.

Which is why Universal Extractor, something I stumbled across today whilst looking at how you install ActivStudio 3 across a network, was an amazing find. It supports loads of different installer types, and is very easy to use.

Universal Extractor | LegRoom.net

Oh, and if you want to know how to roll ActivStudio 3 out on your network, here is an excellent guide.

I run several Drupal-based websites, and today one of them stopped allowing unregistered users access to new content, for seemingly no reason. One solution I found was to upgrade from v5.6 (which I was running) to v5.7, but that didn’t work.

It turns out there’s a really simple fix that worked for me. I just ran the following SQL command:

INSERT INTO node_access VALUES (0, 0, 'all', 1, 0, 0);

I renamed “node_access” to include the Drupal table prefix (you may not have one), and of course I backed everything up beforehand, but it worked.

Why? Well, it seems that user ID “0″ (that is, “anonymous user”) had gone walkabout from the database table. Where it went, I don’t know, but that command restores it. As always, your mileage may vary.