We have been experiencing intermittent publishing issues across all our sites (70 plus) for a while with published content appearing and disappearing from public view whilst always being visible in the admin area. The material eventually appears properly but only after an indeterminate number of days.
We were operating under the impression that data was being cached by the umbraco.config file itself and not realising that this is created on app_start purely to allow an in-memory copy to be made in the IIS AppPool that the site(s) run in. When the penny dropped and we realised that multiple instances of the AppPool would have different versions of the cached data we checked our AppPool settings and found we were running with 2 instances.
We reset this to one instance and all our sites now save, publish and update the cache perfectly. At some point in the last couple of months somebody changed the AppPool settings, probably with good intentions without they or us realising the implication. Because we didn't get immediate user feedback other than on a couple of sites we didn't connect the two events.
Even if you haven't got reported issues I'd suggest checking that you only have a single instance of any AppPool running. In IIS look at the properties of all the AppPools running and on the Performance tab under Web Garden check you only have one instance selected.
Hope this helps somebody else as we have been going mad trying to find it.
Content publishing issue, possible resolution
We have been experiencing intermittent publishing issues across all our sites (70 plus) for a while with published content appearing and disappearing from public view whilst always being visible in the admin area. The material eventually appears properly but only after an indeterminate number of days.
We were operating under the impression that data was being cached by the umbraco.config file itself and not realising that this is created on app_start purely to allow an in-memory copy to be made in the IIS AppPool that the site(s) run in. When the penny dropped and we realised that multiple instances of the AppPool would have different versions of the cached data we checked our AppPool settings and found we were running with 2 instances.
We reset this to one instance and all our sites now save, publish and update the cache perfectly. At some point in the last couple of months somebody changed the AppPool settings, probably with good intentions without they or us realising the implication. Because we didn't get immediate user feedback other than on a couple of sites we didn't connect the two events.
Even if you haven't got reported issues I'd suggest checking that you only have a single instance of any AppPool running. In IIS look at the properties of all the AppPools running and on the Performance tab under Web Garden check you only have one instance selected.
Hope this helps somebody else as we have been going mad trying to find it.
is working on a reply...