Closed afederic closed 8 years ago
hi @afederic
By checking the screenshot, it seems that authorization was successful, even more uluwatu was able to get permission to query cloudbreak api (for example 3 blueprints are received).
Could you please check if any container reports some error, if you try to login again
cbd logs
hi @lalyos I'm out of office now. I'll upload cbd logs on monday. But I'm quite sure that I didn't see any errors from the other container logs. Thanks
hi Ale,
Hmm i don't see any unusual in the logs ... If you could please join the support channel at hipchat: https://www.hipchat.com/gohd6v0n1 we could have a more interactive troubleshooting.
Fo hipchat: no installation/registration is needed, just a browser.
hi @lalyos I joined. Thanks Ale
hi @lalyos thank you for support. I will resize my CBD Openstack VM following the minimum recommendations (2 VCPU, 4GB RAM) that you gave me in the chat room. I will let you know asap. Thanks Ale
hi @lalyos I have resized the VM: 2 VCPU, 6GB RAM. Started cbd I still get "Client cannot access resource server."
[root@cbd ~]# cbd logs
[...]
cloudbreak_1 | 2015-10-26 15:02:40,357 [main] logStarted:61 INFO c.s.c.CloudbreakApplication - [owner:spring] [type:springLog] [id:] [name:] Started CloudbreakApplication in 59.815 seconds (JVM running for 63.83)
sultans_1 | POST login.do - Client cannot access resource server. Check UAA address.
ambassador_1 | 2015/10/26 15:04:30 172.17.0.13:48906 -> node1.node.dc1.consul.:5433
ambassador_1 | 2015/10/26 15:04:30 172.17.0.13:48907 -> node1.node.dc1.consul.:5433
ambassador_1 | 2015/10/26 15:04:40 172.17.0.13:48912 -> node1.node.dc1.consul.:5433
cloudbreak_1 | 2015-10-26 15:04:48,418 [taskScheduler-1] send:36 INFO c.s.c.s.n.HttpNotificationSender - [owner:spring] [type:springLog] [id:] [name:] Could not send notification to the specified endpoint: 'http://XXX.XXX.XXX.XXX:3000/notifications' Cause: I/O error on POST request for "http://XXX.XXX.XXX.XXX:3000/notifications":Connect to XXX.XXX.XXX.XXX:3000 [/XXX.XXX.XXX.XXX] failed: Connection timed out; nested exception is org.apache.http.conn.ConnectTimeoutException: Connect to XXX.XXX.XXX.XXX:3000 [/XXX.XXX.XXX.XXX] failed: Connection timed out
ambassador_1 | 2015/10/26 15:04:50 172.17.0.13:48915 -> node1.node.dc1.consul.:5433
ambassador_1 | 2015/10/26 15:04:54 172.17.0.14:51422 -> node1.node.dc1.consul.:8089
ambassador_1 | 2015/10/26 15:05:00 172.17.0.13:48921 -> node1.node.dc1.consul.:5433
ambassador_1 | 2015/10/26 15:05:00 172.17.0.13:48922 -> node1.node.dc1.consul.:5433
sultans_1 | POST login.do - Client cannot access resource server. Check UAA address.
[root@cbd ~]# uptime
16:08:27 up 8 min, 4 users, load average: 0.10, 0.34, 0.25
[root@cbd ~]# free -m
total used free shared buff/cache available
Mem: 5807 2238 2808 52 760 3301
Swap: 0 0 0
UPDATE: after some time the previous error disappeared and I'm able to login but the GUI is still loading :-(
[root@cbd ~]# uptime
16:13:25 up 13 min, 4 users, load average: 0.19, 0.19, 0.21
[root@cbd ~]# free -m
total used free shared buff/cache available
Mem: 5807 2588 2435 54 783 2948
Swap: 0 0 0
These is (I guess) the relevant log while the GUI is in the loading state
cloudbreak_1 | 2015-10-26 15:29:47,554 [taskScheduler-1] send:36 INFO c.s.c.s.n.HttpNotificationSender - [owner:spring] [type:springLog] [id:] [name:] Could not send notification to the specified endpoint: 'http://XXX.XXX.XXX.XXX:3000/notifications' Cause: I/O error on POST request for "http://XXX.XXX.XXX.XXX:3000/notifications":Connect to XXX.XXX.XXX.XXX:3000 [/XXX.XXX.XXX.XXX] failed: Connection timed out; nested exception is org.apache.http.conn.ConnectTimeoutException: Connect to XXX.XXX.XXX.XXX:3000 [/XXX.XXX.XXX.XXX] failed: Connection timed out
If the "Loading ..." keeps up, it means that the client side js (angular) doesn't get a response from one of the backend (uluwatu/sultans/cloudbreak/periscope). I suppose uluwatu,sultans, and cloudbreak connection is fine (because in the grayed background you see 3 blueprints)
Pleas with on the developer toolbar in your browser, and check the "network" tab, about which request takes so long.
It might be that you opened up the ports 3000, 3001 so you can reach cloudbreak ui from your browser but the micro services itself also use other ports: 8080, 8089 between each other. So please make sure that the cloudbreak container can curl for example uluwatu.
You can test the connectivity between cloudbreak and uluwatu, by
docker exec -it cbreak_cloudbreak_1 bash
curl -I http://XXX.XXX.XXX.XXX:3000/info
If it happens that you have a PRIVATE_IP
set in your Profile
, please remove it, its not necessary.
hi @lalyos SOLVED! THANK YOU VERY MUCH! :+1: the problem was that the CBD Openstack VM should connect to itself via its PUBLIC_IP (the Openstack floating IP assigned to the VM). I have added a security rule to allow all TCP incoming connections from the PUBLIC_IP and the loading has gone ;-) Do you think I should allow all UDP as well?
Thanks Ale
hi,
You don't need to open UDP.
Attila
Hi, our CBD installation on Centos 7.1.1503 hangs after user login as shown in the following picture
CBD logs from uluwatu shows that
It seems it cannot get a response for account details and permissions. Thanks Ale