I have a three-nodes home Cloudera cluster for practicing which has been working fine until some day some changes were introduced without even noticing it, below is the error message:
org.apache.thrift.transport.TTransportException:
java.net.ConnectException: Connection refused
The front-end error is in Hue I can't see any databases nor can I create a new one; It still shows me query history.
What should I do to narrow down the root cause?
Thank you, any clue is appreciated.
I suggest to check all the services of the cloudera manager in order to confirm whether services are all running properly or not. If only a few services have a problem/connection issues then you will be able to narrow down on the root cause. Your query is quite broad and difficult to suggest anything meaningful from the information provided.
Related
I am using Application Insight for Web Pages(Javascript) and seen that for a particular resource it logs errors constantly because of some extension of the browser. How to avoid such exception caused by another plugins/extension creating load on my server. However, we can't disable(or ask to disable) the extension as its equally important for us.
if anyone out there has encountered this earlier or know the way to avoid exceptions to be tracked which are caused by external plugins.
This error is continuing to occur in my console. I am not refreshing or interacting with the app at all. I am just sitting here, watching. Does anyone know what this error means?
GET http://localhost:3000/sockjs/info?cb=poa8l4xy7q ... sockjs-0.3.4.js:832
If this were on a hosted server, I would guess:
You have websockets turned on for your Meteor app (on by default) but somewhere between Chrome and the webserver there is a component that does not support websockets. So Chrome periodically tries to upgrade the connection to websockets but it never works.
There is a load-balancing issue that alternately connects your client to two different Meteor servers. I would expect this to cause a flash of content though so if you aren't seeing that, its probably not this problem.
Since this is localhost and your app is actually still working I don't know why that's happening but perhaps the above guesses will help you track it down.
EDIT: Are you sure your app is still working when you see these? Can you actually insert/update data and see the changes reflected in the database?
We have developed a website. This will be deployed in IIS in server machine of a network. n number of people will access our website from their allocated pc's. What happens now is, if the browser is left open for an extended period of time, it is getting crashed, then i will have to reload the page to get access to my site. What should i do to overcome this problem?
Please advise on this.
Thanks
Venkat
You havent provided much information here so cant really comment anything specific (for e.g. runtime env, database servers, any specific tech. in use such as TCP/IP, web sockets.
There can be number of things that you could look into.
For the start you could look into resources that are not being disposed properly. For e.g. Database connection, large in memory .NET objects for e.g. file streams. Now this could be something as simple as using statements or exception handling(if .NET).
You could use profiler applications such as MiniProfiler (do google), Again if its Microsoft then you could use Microsoft Load Testing to see where the problem is.
Try using fiddler to see if it shows anything.
If you are using TCP/IP communication for e.g. push notifications from server side using websockets or node.js then you need to investigate if there are any threads/connections not properly closed/disposed.
list is sort of endless.. good luck
I tested MSDN StockTrader5 online and it doesn't work
https://azurestocktrader.cloudapp.net/Quotes.aspx?symbols=s:100;s:101;s:102;s:103
what's the cause ?
It's working now...
My guess is that this might have been a genuine outage - but it's impossible from here to tell whether it was caused by an incorrect application upgrade or by an outage within SQL Azure.
Overall, I've found Azure and SQL Azure to be very reliable, but obviously when a someone comes along and sees an error in an online showcase then it's not a good first impression for them.
As stated in the error message:
Login failed for user 'trade'.
It looks like this application is trying to login to the MSSQL database using a username/password that isn't correct. It looks to be a simple configuration error.
Got reports of a web app going down twice in three weeks. Need to do some root cause analysis. works fine after a reboot. I'm not really an expert in this field.
It is hosted on IIS and Windows 2003.
There is nothing interesting in the event viewer, and IIS logs just show lots of successful GET operations. There is nothing interesting in SQL logs on the remote SQL server it connects to.
I'm not sure how to decipher the IIS log. It just looks like a bunch of successful GET messages with no errors.
I don't think I can really get too much further with root cause analysis track down the cause of the issue?
The only thing you could try to get some real results is this excellent blog by Tess Ferrandez. I think that you will find crash lab very enlightening :)
Take a look at this, it might help you find the app shutdown cause.
Depending on your traffic, twice in three weeks doesn't sound like a lot. The root cause may relate to the fix- if you were able to bring it back up by restarting IIS, it could be a memory leak. If you had to restart the server, it could be a deeper problem.