Site inaccessible due to AuthoriseUpgrade. Umbraco thinks it needs an upgrade but it doesn't.
Hoping someone would be able to assist.
So we have weekly server restarts, we've got IIS service set to (Delayed start) so ideally there's zero disruption. But in very rare cases, users browsing Umbraco sites at that exact time of scheduled restart trigger requests to the Umbraco database when it's not available.
Which results in Umbraco thinking there needs to be an upgrade thus redirects ALL traffic henceforth to the upgrade page. Even if seconds later the Database is actually configured correctly, requests bypass it and just go straight to the redirect.
There's no secondary checks. Just one incorrect check and the site is basically offline until we manually trigger a recheck. (We reupload the web.config basically so it checks the connection string again)
We've been dealing with this manually like so for the past 6 months or so now, each time the site is offline for 7+ hours. Which is disastrous obviously for customers.
But it's getting to the point we're wondering, it seems like we're the only ones experiencing this, right? So what are we doing wrong, how can we either:
A) Get Umbraco to check the connection string more than once?
B) Completely disable the AuthoriseUpgrade page, what would happen?
C) Defer loading of website even more until we're sure the database is accessible?
D) Prevent redirects to upgrade? I'd rather a customer gets a 404 or 503 message if they somehow access the site the 1 minute a week it's offline and restarting!
Any ideas would be greatly appreciated!
--
I would note, we manage some 100+ Umbraco installations. From a wide range of versions. 6.0.5 to 8.4.0. MVC and Webforms!
However, the restart this week, which prompted this forum post and re-investigating our options, it happened to 4 sites at once, all of which run Umbraco 7.3.8 on Webforms. 3 of which are the usual suspects in regards to this happening regularly.
We assume the reason is higher traffic. Rather than a difference in configuration. As those 3 usual suspects are our highest traffic domains. But even then, this error usually happens only once a month, per!
Site inaccessible due to AuthoriseUpgrade. Umbraco thinks it needs an upgrade but it doesn't.
Hoping someone would be able to assist.
So we have weekly server restarts, we've got IIS service set to (Delayed start) so ideally there's zero disruption. But in very rare cases, users browsing Umbraco sites at that exact time of scheduled restart trigger requests to the Umbraco database when it's not available.
Which results in Umbraco thinking there needs to be an upgrade thus redirects ALL traffic henceforth to the upgrade page. Even if seconds later the Database is actually configured correctly, requests bypass it and just go straight to the redirect.
There's no secondary checks. Just one incorrect check and the site is basically offline until we manually trigger a recheck. (We reupload the web.config basically so it checks the connection string again)
We've been dealing with this manually like so for the past 6 months or so now, each time the site is offline for 7+ hours. Which is disastrous obviously for customers.
But it's getting to the point we're wondering, it seems like we're the only ones experiencing this, right? So what are we doing wrong, how can we either:
A) Get Umbraco to check the connection string more than once?
B) Completely disable the AuthoriseUpgrade page, what would happen?
C) Defer loading of website even more until we're sure the database is accessible?
D) Prevent redirects to upgrade? I'd rather a customer gets a 404 or 503 message if they somehow access the site the 1 minute a week it's offline and restarting!
Any ideas would be greatly appreciated!
--
I would note, we manage some 100+ Umbraco installations. From a wide range of versions. 6.0.5 to 8.4.0. MVC and Webforms!
However, the restart this week, which prompted this forum post and re-investigating our options, it happened to 4 sites at once, all of which run Umbraco 7.3.8 on Webforms. 3 of which are the usual suspects in regards to this happening regularly.
We assume the reason is higher traffic. Rather than a difference in configuration. As those 3 usual suspects are our highest traffic domains. But even then, this error usually happens only once a month, per!
Anyone got any ideas?
Server restart happened last night and another 2 sites went 'offline' for 7+ hours. Including our own site. Again, both 7.3.8 Webforms.
is working on a reply...