Yet Another Cycling Forum
General Category => Audax => Topic started by: zigzag on 19 February, 2019, 10:06:33 pm
-
when trying to enter the event (tomorrow) the new website seems to be down. should i wait or go to bed? :)
[image deleted]
the old website doesn't seem to be working either:
(https://uploads.tapatalk-cdn.com/20190219/b78168e093d0f027883e6a7a087d53b0.jpg)
-
Yep, event list page looks fine, but click on any event and get 'Specified event (19-75) could not be found. Please go back and select again.'
:'(
-
Going by that classy sprawl of error messages one of the key database files has gone walkies (or is corrupt).
(It also shows they're using Dapper, which is a database abstraction layer. Trying to talk to a legacy DB with a complex format through an ORM might explain why they made such a meal of getting it to work)
-
Going by that classy sprawl of error messages one of the key database files has gone walkies (or is corrupt).
(It also shows they're using Dapper, which is a database abstraction layer. Trying to talk to a legacy DB with a complex format through an ORM might explain why they made such a meal of getting it to work)
A properly designed Relational database is always going to have that problem, ORM's only easy when you're just using the RDBMS to persist data for an object model... in which case an RDBMS is probably not the right persistence layer...
But hey some people think ORMs save the world and their need to understand what they're using...
-
Maybe the website needs another 50p in the meter?
-
Just logged in to check my results page, page loads, but I have no results. Looks like the DB backend needs a hug :(
J
-
That's annoying. Website still down, so I can't login to buy a new pack of virtual brevets or enter the DIY I was gonna start at 3:30. Nice weather for a 200!
-
Please be patient.
Don't forget that AUK commissioned Charles Babbage to operate the server for the new website. Once in a while it requires some winding to get it going again.
(https://i.pinimg.com/originals/47/2d/f9/472df9bc3a913a1e7978219f6f5557ec.jpg)
-
At first glance this looks quite serious. I'm having difficulty accessing anything in a useful way, the whole filesystem is marked as read-only. A recent data backup looks OK and maybe we'll activate our reserve server, but don't expect anything to happen instantly.
-
Please be patient.
Don't forget that AUK commissioned Charles Babbage to operate the server for the new website. Once in a while it requires some winding to get it going again.
(https://i.pinimg.com/originals/47/2d/f9/472df9bc3a913a1e7978219f6f5557ec.jpg)
You are naughty.
-
Please be patient.
Don't forget that AUK commissioned Charles Babbage to operate the server for the new website. Once in a while it requires some winding to get it going again.
The new website is fine. But it doesn't actually do anything other than look pretty.
I'm being told this might take a while to fix (as FF says).
-
Going by that classy sprawl of error messages one of the key database files has gone walkies (or is corrupt).
Yes, really good of the new website to expose the inner workings of the old one. Not very graceful of them.
-
That's annoying. Website still down, so I can't login to buy a new pack of virtual brevets or enter the DIY I was gonna start at 3:30. Nice weather for a 200!
Cheer up, tomorrow and friday look even better :thumbsup:
-
Just out of interest, is the MySQL server that powers both sites in the custody of ControlF1 or AUKweb?
-
Time to deploy the Audax UK backup procedures for when the website is borked.
(https://www.dropbox.com/s/3gldwtadffpl9nm/audax_backup.jpg?raw=1)
-
Just out of interest, is the MySQL server that powers both sites in the custody of ControlF1 or AUKweb?
The results database is still running on the old server and AFAIK is a basic LAMP setup.
-
All the data is still hosted on aukweb. FWIW I don't think it's the database that is the fundamental problem here. It's in hand but really my best advice until further notice is "try again tomorrow".
-
as far as DIY rides go, hopefully one can simply email one's intended route and date to local DIY man, complete the ride, and paperwork can be completed and brevets purchased when site is back up?
-
I don’t enter DIYs until just before I ride them, and didn’t want to email or ring Mike Kelly at 3am if it wouldn’t be validated. Out on a club ride now.
-
as far as DIY rides go, hopefully one can simply email one's intended route and date to local DIY man, complete the ride, and paperwork can be completed and brevets purchased when site is back up?
I intend to do this if the site is not working properly this evening. Fortunately I'm riding a route which I've used before and the local DIY man (Manotea OTP) will be familiar with. I bought my virtual brevets yesterday :smug:
Going by that classy sprawl of error messages one of the key database files has gone walkies (or is corrupt).
Yes, really good of the new website to expose the inner workings of the old one. Not very graceful of them.
The security spooks where I work wouldn't like the call stack trace and would be horrified that the name of a datafile had been revealed (even if a relative path). It's interesting to see they're using that dotNOT async stuff - well down with the kids good modern practice.
-
What gets me about the publicly visible stack trace in the error messages is that you have to actively turn that on...
On the other hand it let me go into work mode and find out that mysql error 30 is a read-only file system error....
-
Here's a 'paper' entry form for anyone who needs it. Not quite the current version but will suffice for now.
https://drive.google.com/file/d/1YZYwYH9pIT5v3AqrxJg2HvRUoQm9V7aR/view?usp=sharing
-
The security spooks where I work wouldn't like the call stack trace and would be horrified that the name of a datafile had been revealed (even if a relative path).
In fact I'd quite appreciate it if the OP could remove that image from his post. The error message is there I assume because the new website is still in alpha/beta.
-
Id quite like to download the rider start list for my event on saturday!!
-
Showing stack traces for remote requests only happens if you change the 'custom errors' setting from its default of RemoteOnly to Off. Quite bad practice to set RemoteErrors to Off for a production/public-facing website, even if it's only in alpha/beta.
(Whilst I'm at it it also really gets up my goat if they supply a custom error but it contains any derivation of the word 'oops' or 'whoops' or 'whoops-a-daisy' etc....) ::-) ::-) ::-)
-
Well hopefully it will be fixed by the weekend; given it's going to be scorchio (for February) there is likely be a last minute flurry of entries; myself included.
-
www.aukweb.net is working again.
Andy Corless
-
...and I've just entered a mandatory route DIYxGPS successfully :thumbsup:
-
www.aukweb.net is working again.
Andy Corless
Limited access - you might get what you want, you might not.
Some areas are prompting you to return in a couple of days
-
just arrived home from the above audax, so could not see the request to delete the image earlier.. i rode to the start to see if i can ride and thankfully the organiser kindly accepted entry on the line and had spare brevet cards; a day well spent with 269k in the legs.
-
thankfully the organiser kindly accepted entry on the line and had spare brevet cards
Maybe my loss was your gain!
-
thankfully the organiser kindly accepted entry on the line and had spare brevet cards
Maybe my loss was your gain!
could have been, there were a dozen or so of dns cards on the table. i hope all is well!
-
Seems to be fixed, I can log in.
J
-
It's interesting, the complete silence from the board on the matter. Not a whisper.
-
It's interesting, the complete silence from the board on the matter. Not a whisper.
Early days yet, I'm still awaiting a reply to a query sent Autumn 2017 .
-
It's interesting, the complete silence from the board on the matter. Not a whisper.
Not sure what you are expecting here. This was an issue with the old server that in the slow process of being replaced by the new site (as discussed ad nauseum), if anything it reinforces their decision to proceed with the IT refresh.
-
It's interesting, the complete silence from the board on the matter. Not a whisper.
Not sure what you are expecting here. This was an issue with the old server that in the slow process of being replaced by the new site (as discussed ad nauseum), if anything it reinforces their decision to proceed with the IT refresh.
Just a statement, something like "sorry, we're working on it". An indication that someone had noticed,
-
There were a couple of tweets[1] from @AudaxUK (https://twitter.com/AudaxUK/status/1098153858768011264)
But in general with these things, it's a balance between trying to solve the problem and communicating what's going on. There are strong arguments that time spent communicating prolongs the lack of resolution, and that you sometimes need to get a long way towards solving the problem before you can make realistic estimates. The right balance is only ever clear in hindsight.
[1] Obviously these things are best done out-of-band: Even if it were practical to modify the website(s) to announce the issue, who's to know they aren't going to have to bring the whole thing down in order to resolve it? Twitter is a pretty good platform for this sort of thing, as it's easy to find and follow in real-time, easy to update and doesn't suffer the walled-garden problems of Facebook. I don't know if the AUK forum was affected by this issue, but conceivably there are circumstances where it might be (eg. by a hosting or DNS problem).
-
Not sure what you are expecting here. This was an issue with the old server that in the slow process of being replaced by the new site (as discussed ad nauseum), if anything it reinforces their decision to proceed with the IT refresh.
It does highlight that we're still 100% dependent on the old site staying up, and will be for several more years, possibly indefinitely, yet there's no sign of any plan or money for its upkeep and maintenance.
-
Obviously these things are best done out-of-band: Even if it were practical to modify the website(s) to announce the issue, who's to know they aren't going to have to bring the whole thing down in order to resolve it? Twitter is a pretty good platform for this sort of thing, as it's easy to find and follow in real-time, easy to update and doesn't suffer the walled-garden problems of Facebook. I don't know if the AUK forum was affected by this issue, but conceivably there are circumstances where it might be (eg. by a hosting or DNS problem).
+1
-
There were a couple of tweets[1] from @AudaxUK (https://twitter.com/AudaxUK/status/1098153858768011264)
But in general with these things, it's a balance between trying to solve the problem and communicating what's going on. There are strong arguments that time spent communicating prolongs the lack of resolution, and that you sometimes need to get a long way towards solving the problem before you can make realistic estimates. The right balance is only ever clear in hindsight.
I know about the tweets cos I sent them.
The only people I know have been involved are FF, Terry (both IT delegates) and the one who reported it (me).