Some having issues using AFF 11/11/2010

[serfty]

Reports on FT of some having issues connecting to AFF:

• OT - Australian Frequent Flyer down (9 April) - FlyerTalk Forums

 

[JohnK]

Yes AFF appeared to be extremely slow from around 10:30am through to around 1:00pm.

No issues since....

 

[admin]

Reports on FT of some having issues connecting to AFF:

• OT - Australian Frequent Flyer down (9 April) - FlyerTalk Forums

No issues that I'm aware of. The load on the server was quite high this afternoon (due to The Gazette newsletter going out), but nothing unusual.

Regarding the issue on FT: Looks to me that the poster (zan5hin) was viewing a sitemap (eg. http://www.australianfrequentflyer.com.au/community/sitemap/f-72.html) rather than the actual site. This is easily rectified by going back to the home page and then navigating to where you want to go.

If anyone is having similiar problems - and its not the sitemap - please PM me and I will investigate.

 

[jsd]

Is anyone seeing the Homepage like this when they load up AFF, it fixes itself when you login but its quite annoying :-|

Fairly confident that my laptop is not the problem as I was having the same issue with 2 different PCs running IE today

 

[samh004]

Is anyone seeing the Homepage like this when they load up AFF, it fixes itself when you login but its quite annoying :-|

Fairly confident that my laptop is not the problem as I was having the same issue with 2 different PCs running IE today

You’re right there is a problem loading the site when not logged in.

Reporting your post to admin.

 

[straitman]

You’re right there is a problem loading the site when not logged in.

Reporting your post to admin.

Strange, I logged off and opened and closed pages and everything appeared to work normally for me.

 

[samh004]

Strange, I logged off and opened and closed pages and everything appeared to work normally for me.

I’m getting the issue with Chromium and Chrome, but not Firefox or Safari.

 

[admin]

Very odd... Something similiar happened to me a couple of weeks ago. It self-corrected, and I haven't come across the problem since. Will need to do a bit more investigation.

Thanks for in info and screenshot

Is anyone seeing the Homepage like this when they load up AFF, it fixes itself when you login but its quite annoying :-|

Fairly confident that my laptop is not the problem as I was having the same issue with 2 different PCs running IE today

View attachment 1432

 

[drron]

I have had pages loading slowly from time to time but couldnt access AFF at all this morning from 0630-0730 Tassie time.

 

[samh004]

I have had pages loading slowly from time to time but couldnt access AFF at all this morning from 0630-0730 Tassie time.

I had troubles last night around 0200

But just went to bed instead.

 

[maggies1]

No AFF at all this morning. Could not connect from 0500am, for a few hours.
maggies1

 

[oz_mark]

I occassionaly get a message (pop up box if on the home page, or text on a page if elsewhere), about a security token missing.

 

[drron]

I had troubles last night around 0200

But just went to bed instead.

That's a good boy!

 

[admin]

There are a few (unrelated) issues discussed on this thread. Let me explain what's happened over the last couple of weeks:

1. Last week non-paid users would have experienced very slow response times. The problem was with the server we use to display ads. AFF was waiting for the ads to become available before it could display a page. So, to members who see ads, it appeared as though AFF was slow. That issue was resolved last Friday. The ad server was upgraded and the Ad-serving software re-optimised for the new hardware.

2. There was an unrelated issue in the early hours this morning with the main AFF server. This is a different server to the Ad server, mentioned above. We own the hardware for the AFF server which is located in a secure Data Centre near Sydney. (The ad server is on a smaller leased machine in the USA.) Evidently the Data Centre had a "switched PDU trip" - whatever that is :?: and all servers on the rack were down from about 2:30am till 8:30 am. I am obviously not happy with the length of time it took to resolve the problem, and am exploring alternatives.

3. The other issues raised on this thread are not hardware related. I am unable to consistently replicate the problem so its difficult to diagnose. Doesn't appear to be a major issue. I suspect its cookie-related on some browsers under certain conditions.

So thats what's been happening in "admin world" this last week!

Rest assured that we always try to provide the best possible service to our members - I believe things are usually pretty good, but we do get the odd bad week! Apologies for any inconvenience.

 

[NM]

Evidently the Data Centre had a "switched PDU trip" - whatever that is :?: and all servers on the rack were down from about 2:30am till 8:30 am. I am obviously not happy with the length of time it took to resolve the problem, and am exploring alternatives.

PDU = Power Distribution Unit.

Data centre design best practice (known as DCD-101) is to have two power feeds into each rack, sourced from independent PDUs, and each device in the rack should be dual-powered with a feed from each of the independent power circuits. any devices that do not support dual power should be established in redundant pairs sourced from the two different power feeds.

A single PDU trip should not result in any equipment failure, just a loss of power supply redundancy until the downed PDU feed is rectified.

The most common cause of a tripped PDU is human error during some form of maintenance activity within the data centre, either by overloading a circuit (did not do the calculations correctly) or turning off the wrong circuit.

 

[straitman]

PDU = Power Distribution Unit.

Data centre design best practice (known as DCD-101) is to have two power feeds into each rack, sourced from independent PDUs, and each device in the rack should be dual-powered with a feed from each of the independent power circuits. any devices that do not support dual power should be established in redundant pairs sourced from the two different power feeds.

A single PDU trip should not result in any equipment failure, just a loss of power supply redundancy until the downed PDU feed is rectified.

The most common cause of a tripped PDU is human error during some form of maintenance activity within the data centre, either by overloading a circuit (did not do the calculations correctly) or turning off the wrong circuit.

NM,

On behalf of LW, myself and all the non computer geeks here I say:

Yeah, what ever you said :!:

 

[Danger]

admin, from today I've stopped receiving automatic and instant email notifications of updates to threads I've subscribed to. All was fine overnight but not from my first logon to the site this morning (around 9am WST). My email address has not changed, I haven't made any changes to my notification preferences and I'm receiving other emails fine (including notification emails from FT).

 

[Flashback]

admin, from today I've stopped receiving automatic and instant email notifications of updates to threads I've subscribed to. All was fine overnight but not from my first logon to the site this morning (around 9am WST). My email address has not changed, I haven't made any changes to my notification preferences and I'm receiving other emails fine (including notification emails from FT).

Ditto, and on that note it's time to head to work

 

[admin]

You should now be getting email. There was a problem when the server recovered from the outage earlier today - thats now been rectified. Thanks for bringing this to my attention.

admin, from today I've stopped receiving automatic and instant email notifications of updates to threads I've subscribed to. All was fine overnight but not from my first logon to the site this morning (around 9am WST). My email address has not changed, I haven't made any changes to my notification preferences and I'm receiving other emails fine (including notification emails from FT).

 

[Flashback]

You should now be getting email. There was a problem when the server recovered from the outage earlier today - thats now been rectified. Thanks for bringing this to my attention.

Can confirm I just received one.