|
The article has been removed.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Thanks, can you kick out me from CP ?
Thanks
|
|
|
|
|
I've noticed that Web12 displays the white bar at the top with a bigger font than the other servers. Not really big, just about 10pt or so.
I haven't noticed it on other servers than Web12.
Thought I just wanted to report this one. Not anything critical, but still.
Kristian Sixhoej
"I have noticed even people who claim everything is predestined,
and that we can do nothing to change it, look before they cross the road." - Stephen Hawking
|
|
|
|
|
Web12 has left the building.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
You must have quite a lot of servers sitting in the parking lot going by recent posts!
|
|
|
|
|
Just what I was thinking - yet another server thrown out of the window.
Kristian Sixhoej
"I have noticed even people who claim everything is predestined,
and that we can do nothing to change it, look before they cross the road." - Stephen Hawking
|
|
|
|
|
Yes, and the ones remaining are being very well behaved.
Each time a server fails we take it out of the rack and show it to all the other servers. We explain how this particular server was asked to do a very simple job but it failed, and how servers that can't perform their duties are sent...elsewhere. The other servers are given a chance to digest this, then, with no ceremony, we unplug (without shutting down) the failed server and wheel it out of the room. Slowly.
Our server fail rate has dropped dramatically.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
All of that done by hamsters, right?
“Follow your bliss.” – Joseph Campbell
|
|
|
|
|
Chris Maunder wrote: and how servers that can't perform their duties are sent...elsewhere.
That'll be a server farm, won't it? Where they can play all day...
(Where's the emoticon of a puppy with massive eyes?)
Iain.
I have now moved to Sweden for love (awwww).
|
|
|
|
|
er...yes, that's right. It'll be happier there, really.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
is there a way to direct a request to a particular server? can be add something to a URL so it will be say web21 who should deal with it? That would ease the anamnesis of some problems.
|
|
|
|
|
Luc Pattyn wrote: is there a way to direct a request to a particular server
Internally, for us, yes; externally, for you, not reliably for a bunch of really annoying reasons I won't go into.
Are you thinking that, in order to test and replicate, you would just type in "web21.codeproject.com" to confirm it was a particular server? If so, then yes, that can be handy, but one of the issues we found when we had this enabled previously was that a lot of people decided that a particular server was more reliable than other servers so hit that one directly instead of letting the load balancer deal with this. This, ironically, made that particular server suffer more wear and tear.
BTW: anamnesis is my new word of the day.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Chris Maunder wrote: web21.codeproject.com
you could enable that for a selected audience, say gold organizers, or SiteBuilders.
Chris Maunder wrote: anamnesis is my new word of the day
really? in your line of business, I bet you need it constantly. Good anamnesis yields good diagnosis.
|
|
|
|
|
one option to allow us to do some per server testing without flogging your servers to death would be to have all links from webNN.codeproject.com point to www.codeproject.com. Having to copy each link and manually edit it in the address bar would stop load balance abuse while still leaving the people with too much time free to try and troubleshoot where problems are coming from...
3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
I voted for an article to win a few days ago.
a) *I* voted for it, so why wasn't the contest closed and my favoured article declared the winner? Don't you know who I am?
b) I was curious how it was doing, so I went to the contest page - but it only showed me the up to date voting after I tried to cheat and voted again. It does not feel right re-voting, even though you detect my attempts at fraud. If you can detect that I'm dishonest, can't you detect that I've already voted and show the graph instead of the voting list?
c) I fully support not showing the voting progress until after a vote has been placed. Seems like a good way to encourage blind honest voting.
Cheers,
Iain.
I have now moved to Sweden for love (awwww).
|
|
|
|
|
a) Clearly we have a bug. I'll add it to a list somewhere.
b) We could, though the feeling at the time was to just show the voting form again to allow someone to re-cast their vote. I think what we need is a "AllowRevoting" flag that, if set to true, shows the voting form; otherwise just the results.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Chris Maunder wrote: I'll add it to a list somewhere
This is going to be my new answer to everything
DaveIf this helped, please vote & accept answer!
Binging is like googling, it just feels dirtier. (Pete O'Hanlon)
BTW, in software, hope and pray is not a viable strategy. (Luc Pattyn)
|
|
|
|
|
Chris Maunder wrote: a) Clearly we have a bug. I'll add it to a list somewhere.
I guess it's better than doing nothing about it.
“Follow your bliss.” – Joseph Campbell
|
|
|
|
|
|
Off late, i have observed that Zero or anything below zero is no more shown in the reputation points.
|
|
|
|
|
zero rep never got shown; negatives have been there and are no longer. Luckily we still have the rep graph.
|
|
|
|
|
You don't have time or interest to keep enhance articles you wrote a long time ago, sometimes the code is out-dated and should not be used. It would be nice to allow authors mark/flag which of his/her articles are "obsolete", etc.
|
|
|
|
|
I think this is a good idea. Obsolete articles would be those that are too out of date to really be useable, but which still might contain useful techniques or otherwise be good reference. The method for flagging an article obsolete could also give the author an option of pointing to a more up-to-date version.
|
|
|
|
|
Yep, now I see that my first articles was really awful and a quality of source code was terrible. But I can't just delete them if there are people (or even bots) who have bookmarked them.
Greetings - Jacek
|
|
|
|
|