Paw crashed on me today and when I reopened it all of my save requests and environments were gone. Where is that data stored and is there any hope of recovering it? Thanks
Related
Before I launch a Realm app that lots of people rely on, I'm trying to get acquainted with all the possible Realm Object Server failure scenarios so that I'm prepared. :)
Let's say I have the ROS deployed successfully on an Ubuntu VPS and everything works great. Then suddenly my VPS provider shuts down and I have to migrate immediately to another.
I push a client app update with a new realm SyncConfiguration with a new server URL and realm URL and it points to a freshly-installed instance of the ROS with no realm data.
When the users' apps connect to the new server and re-authenticate, what happens to their data when they sync?
All their local data syncs and pushes up to the ROS and everything resumes like the situation never happened.
The new ROS overwrites the local realm with no data.
Something else
I know I can do server-side backups (and I will), but I'm just trying to anticipate what a server migration would look like.
Thanks!
You will receive an error called Client Reset. This happens when the server and client disagree about the history of the Realm. It a Client Reset happens, it will backup the local file to another location on the device, after which the original file gets deleted so the server state can be synced.
You can read more here: https://realm.io/docs/realm-object-server/#client-recovery-from-a-backup
How to handle it depends a little on what binding you are in, but the overall concept is the same. This is Swift way:https://realm.io/docs/swift/latest/#client-reset
I am accessing a Drupal Views feed through xmlrpc. The script has worked in the past and my goal today was solely to access another feed. In theory, there was nothing to do except to change the name of the feed. The endpoint had not changed, my domain had not changed, I can log in to the remote site so my user credentials there are valid.
I am scratching my head as to what may have changed. Is there an obvious question that I have missed? What could have changed on the Drupal end that I should be taking into account?
I can also get a session id for an anonymous user okay.
The failure comes during the complicate authentication (that has worked in the past).
Any suggestions?
Thanks.
Ah... if anyone else has the same problem, as I worked through my script, printing out its effect at each line, I came across a comment I had made when I wrote it.
Make sure the client and the remote are on the same time, preferably the time provided by www.time.is.
My PC was running a minute slow. The detafult Resynchronise on Windows 7 runs at 1am on a Sunday. Change that to a more sensible time.
And for an immediate fix, change the PC time to within a few seconds of www.time.is.
That was the problem. Authenticated login uses a time stamp. It the remote server regards your time as too inaccurate, it will reject your login. Make sure the client is running with an accurate clock.
We have an ASP.Net web application on IIS7 that is used to upload Excel files and then load them into a SQL database by running jobs on the SQL server. The app will wait until the job completes then show the user a message. Due to some larger files being used the app is throwing the error below.
Network Error (tcp_error)
A communication error occurred: "" The Web Server may be down, too
busy, or experiencing other problems preventing it from responding to
requests. You may wish to try again at a later time. For assistance,
contact your network support team.
The app uses an asp:View to progress from various steps. I have tried to bump the session timeout and httpRuntime executionTimeout values to account for how long the job takes to run but it does not appear to have any effect. I know the job completes but the app isn't showing that feedback to the user. I think the error is thrown as the app hits the logic to display the user the view showing all the final messages.
I can only guess that a) there is another setting I'm not aware of for timeout, b) another config file setting is overruling my web config setting for the app, or c) the asp:View is counting all the various steps as one long process and not reseting the "clock" as each step is completed.
As I said, the file upload fine, and the job completes fine, the app just can't advance to that last step where it shows the user the view upon the end. Any ideas on what I can look for to fix this issue? My only other option would be to rewrite the app to not wait for the job to finish and handle notifying the user some other way.
Update 1
After further testing it appears the error is from the ASP.Net custom code we created that does a SQL bulk copy and not the running of the SQL job. The current test runs around 220 seconds testing locally but causes a timeout on a test server.
Update 2
After more research I'm inclinded to think user pevgeniev is correct and this is just a limiting factor of the browser. The only thing that prevents me from marking this as answered is I don't know why file uploads don't appear to have the same issue.
If you're getting this error in the browser, than the timeout is on the client side, and there isn't much you could do server side. As you've suggested, you could rewrite the app, so that it polls for the result from the client, rather than expecting to finish the task in a single request.
I am having a lot of problems with the behaivour of the socket (not closing or opening correctly) used in a webapp to communicate with another program running on a server. The problems I hava found is that the socket either ends up being opened or closed various times during the same request or it does not ends even when I kill the active session.
Currently I save the socket in a Session Variable and I check whenever it should be created or if the value of the variable should be referenced, although I am still having a lot of problems.
Have you faced something similar and would you share some of your experience with me?
Also at first I tried to make it Static although I had even more problems with it.
Thank you.
You put your active open sockets in cache or application object in dictionary or whatever suits you provided by asp.net. Before opening new port you check if it is already open by looking in dictionary stored in application or cache. Remove the entry from dictionary when conection is lost.
Imagine this szenario: Web-Server and WebDAV-Server using same cookie-domain.
WebServer sets some authentication cookie.
When starting an WebDAV-Session via special-link on the Web-Server,
WIN7 Microsoft Mini-redirector can read the cookie
set by Web-Server fine and starts an WebDAV-Session
on WebDAV-server using this cookie.
But after the first DAV-call, the cookie seems to be read-only
to both DAV-Server and mini-redirector.
DAV-Server cannot update this cookie
DAV-Server cannot delete (expiry in the past) the cookie
Mini-redirector keeps sending the old values
(even, if Web-Server has updated its values in between)
Anybody got informations
where mini-redirector stores its cookies
how to avoid this behaviour?
Thanks, Karl
I had the similar problem. On the client you should know when the server is updating the cookie. When you know server may reset the cookie, call wnetcancelconnection2 and clear of the the existing webdav connection and call wnetaddconnection2 to create the connection again. But give some time between these two calls atleast a minute, i guess it takes a while to clean up the existing cache.
In my application when i knew the server is going down or my application exiting or going to ask for new session, i call the wnetcancelconnection2 and wnetaddconnection2. It worked for me. Hope this helps.