Amazon Web Services Blog

If you signed up with the National Do Not Call Registry when the service opened its doors in 2003, it’s time to re-register (registration expires after five years). You can re-register anytime, so go ahead and refresh even if your five years aren’t up. [via]

Windows/Mac/Linux (All platforms): Free, cross platform application SvnTimeLapseView downloads every revision of a file from your Subversion repository and scrolls through revisions with a simple slider, highlighting differences in blue. If you're not familiar with the benefits of version control (especially for people who live in text), check out our guide to setting up a home subversion server for a better idea of what it does and whether Subversion might be useful for you. If you decide Subversion is up your alley, Windows users should also check out how to use TortoiseSVN with Subversion. SvnTimeLapseView is free, works on Windows, Mac, and Linux (or any platform that runs Java).

We built a custom Centos5-x86_64 instance and installed the 2.6.16.33 kernel source and kernel modules (for x86_64). It boots up fine if /dev/sda2,3 aren't in fstab but if we add them as outlined in the ec2 manual, we get the following output (from dmesg):

device-mapper: 4.5.0-ioctl (2005-10-04) initialised: dm-devel@redhat.com
MC: drivers/edac/edac_mc.c version edac_mc  Ver: 2.0.0 Aug 15 2007
NET: Registered protocol family 2
Registering block device major 8
 sdb: unknown partition table
 sdc: unknown partition table

A little later in the boot process, there are other filesystem related errors (not sure if they're connected):

md: Autodetecting RAID arrays.
md: autorun ...
md: ... autorun DONE.
kjournald starting.  Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.
VFS: Mounted root (ext3 filesystem) readonly.
md: Autodetecting RAID arrays.
md: autorun ...
md: ... autorun DONE.
EXT3 FS on sda1, internal journal
loop: loaded (max 8 devices)
kjournald starting.  Commit interval 5 seconds
EXT3 FS on loop0, internal journal
EXT3-fs: mounted filesystem with ordered data mode.
Buffer I/O error on device loop0, logical block 206866
... (repeats multiple times) ...
Aborting journal on device loop0.
EXT3-fs error (device loop0) in ext3_dirty_inode: IO failure
EXT3-fs error (device loop0) in ext3_ordered_commit_write: IO failure
ext3_abort called.
EXT3-fs error (device loop0): ext3_journal_start_sb: Detected aborted journal
Remounting filesystem read-only
... (more Buffer I/O errors) ...
__journal_remove_journal_head: freeing b_committed_data
.. repeats for the rest of the log except for a "journal commit I/O error"


Any ideas what we're doing wrong?

Thanks,
Colin

The default Windows Explorer sidebars—generally either the System Tasks and Other Places view or the file system tree—are handy ways to navigate your filesystem, but if you're an ex-Mac user or you would prefer a Mac-like favorites sidebar that lets you define what folders you want quick access to via drag and drop, weblog UneasySilence's simple guide to mimicing OS X's Finder Sidebar in Windows Explorer is for you. By installing a program called Finder style and performing a simple registry tweak, you can turn your sidebar into an attractive, customizable point of access to all of your most used folders.

We've covered a similar method in the past using freeware application FindeXer, but the UneasySilence method yields much more attractive and usable results. When you're finished with the tweak, Finder Bar will always show up as the default sidebar when you open Explorer (something that never worked quite right for me with FindeXer).

While Explorer's Favorites sidebar does serve a similar purpose, Windows doesn't make it easy to set it as your default sidebar, and it honestly just isn't all that attractive.You can customize the Finder bar with your favorites, reorder folders, and set custom names and icons for any shortcut. If you're not happy with the Finder Bar, UneasySilence provides a second registry file to restore the default Explorer sidebar.

Hello,

If I do a search for a CD on amazon UK using ECS, for example:

http://webservices.amazon.co.uk/onca/xml?Service=AWSECommerceService&Operation=ItemLookup&AWSAccessKeyId=[myID]&ItemId=B00005NUZO&MerchantId=All&ResponseGroup=OfferFull,ItemAttributes&OfferPage=3&Condition=All&Version=2005-10-13

(n.b. I've removed my id from the above url)


When viewing this item on Amazon.co.uk's site I can see that amazon's own price for this CD is £5.47. Now, the only offer at this price in the XML returned by the query above has a merchant ID of A3P5ROKL5A1OLE - Amazon UK's merchant ID, which is exactly what you'd expect. However, if when viewing this item directly on amazon's site I add it to my cart and go through the checkout process, I can see that the shipping on this item comes to £1.24 - which according to the shipping scheme outlined here ( http://www.amazon.co.uk/gp/help/customer/display.html?nodeId=11072961) is the rate you'd expect for a single CD supplied by Indigostarfish.com, not the rate you'd expect if purchasing directly from amazon uk (which I calculate to be £1.45). So presumably this CD is in fact being supplied by Indigostarfish, despite the fact amazon's merchant id is given.

Is there a way I can determine programmatically who's supplying a CD offered via an ECS query?

Hope that makes sense!

Any comments or suggestions are gratefully received.

Thanks, Luke

Hello.

I'm trying to send email from my instance using PHP/Apache (or even sendmail directly), and for some recipients it always fails.

It works fine for yahoo mail, and gmail.

However, I haven't been able to get it to work for several other domains.

Here is an example log from an attempt I just made (by calling sendmail directly with logging, but the same thing happens with an apache page):

Oct 19 12:40:53 domU-12-31-36-00-09-94 sendmail[4074]: l9JGer8g004074: to=silikon@alumni.com, ctladdr=root (0/0), delay=00:00:00, xdelay=00:00:00, mailer=relay, pri=30015, relay=[127.0.0.1] [127.0.0.1], dsn=2.0.0, stat=Sent (l9JGerB2004077 Message accepted for delivery)   Oct 19 12:41:00 domU-12-31-36-00-09-94 sendmail[4079]: l9JGerB2004077: to=<silikon@alumni.com>, ctladdr=<root@domU-12-31-36-00-09-94.z-1.compute-1.internal> (0/0), delay=00:00:07, xdelay=00:00:07, mailer=esmtp, pri=120385, relay=alumni-com.mr.outblaze.com. [208.36.123.17], dsn=5.1.1, stat=User unknown   Oct 19 12:41:00 domU-12-31-36-00-09-94 sendmail[4079]: l9JGerB2004077: l9JGf0B2004079: DSN: User unknown

Is this something to do with DNS relating to my virtual server? Are the cloud servers black listed by some ISPs? Has anyone else seen this? Where does "domU-12-31-36-00-09-94.z-1.compute-1.internal" come from and how can I change it?

Last night and this morning I've seen a number of my instances launched take a particularly long time to move into the "booting" phase... One instance that seems to be have been sluggish is i-a506eecc. Are there any known problems with any of the nodes on the grid? Am I losing the node lottery here? Not all of them are slow--they usually take 4-5 minutes to provision and start booting. Some of them, though, are taking 10-15 minutes.

Thanks!

-Tim

Is there a way to set the Cache-Control header on images/files uploaded to S3? I have tried setting it but it converts it to x-amz-header-cache-control which obviously doesn't actually cause the image to be cached when accessed in the browser!

Please help!

thanks
Marcus