|
|
| Subject: | zproxy/store_answering failing |
| Author: | [Not Specified] |
| Posted: | 2015-10-19 14:32 |
I have deployed 5.0.7 multiple times since its release last week, only to have the same results over and over again: I can't even get the Zenoss.core application deployed successfully.
Both instances of MetricShipper fail the health check with store_answering.
Zenpython, zenhub, and Zope are all failing the health check with zproxy_answering.
I have followed the instructions exactly as documented, and this has to be at least my 50th deployment with Zenoss 5.0.x., so I know I have not made any errors in the deployment process. The services simply do not pass the health check, and as a result, the core application doesn't work. This isn't the first time I have come across these issues, as I've seen them in every version since 5.0.3 (my first deployment - actually the only stable deployment I've had, but since it couldn't monitor IPv6, it was worthless to me).
This actually seems to be a random occurrence; one deployment may result in this issue, while another instance configured exactly the same could work, hence why I've deployed 5.0.x over 50 times already.
Is there a fix to this issue, or at least a workaround What even are store and zproxy, and what could be causing them to fail so often I have found no documentation whatsoever regarding these services/processes.
| Subject: | 5.1 Projected Release Date? |
| Author: | [Not Specified] |
| Posted: | 2015-10-19 14:45 |
To say the least, I am very frustrated with Zenoss Core 5 so far. With all of the bugs and instability I've encountered, Zenoss Core 5 is clearly not production ready. Nor does it seem to be in the near future, with releases coming every month to fix bugs that shouldn't have been issues if this was truly production ready in the first place (the search function doesn't work until 5.0.7 IPv6 monitoring doesn't work until 5.0.6 what a joke). Not to mention with each fix comes more breaks (Can't change host IP in 5.0.6, when I did it many times in 5.0.3 http://www.zenoss.org/forum/135566). At the very least it would be nice if the application actually worked on initial deployment in the first place.
Anyone know when is 5.1 projected to be released
I am at the point of giving up on any 5.0.x releases due to the countless issues I've come across. I won't trust any of these to successfully monitor a production environment until I can deploy something that I have proven to be stable. It is a waste of my time to keep deploying these versions if they continue to fail to meet production standards, or don't even start on initial deployment.
| Subject: | Hi, |
| Author: | [Not Specified] |
| Posted: | 2015-10-20 05:14 |
Hi,
I must say, as a potential new customer for this product, this sounds very disconcerting at least. I'm in the process of trying 5.0.7 out, and wouldn't know what to do if I encountered these showstoppers.
/melgaard
| Subject: | same issue as OP |
| Author: | [Not Specified] |
| Posted: | 2016-04-19 04:29 |
Hi ,
We have amonitoring system usingZen Packs based onZenoss 5.0 and with the exact same OS and product package we get the above issues on one site, while on two other sites with the same OS/product package we don't get that issue.
After install the below scenarios occur:
"""
Both instances of MetricShipper fail the health check with store_answering.
Zenpython, zenhub, and Zope are all failing the health check with zproxy_answering.
"""
This error occurs before any ZenPacks are loaded / any devices/events are configured.
After Multiple restarts, OS re-installs, package re-installs the issue still exists.
The System Requirements are met and exceeded on this server.
Has anbody found out why this happens on some systems and not on others with the same setup/packages
An upgrade is not possible in the new few months and a lot of re-development and testing is needed, so any help on wht these errors occur on some installs would be great or if anybody found a workaround for them.
John
| Subject: | same issue as OP |
| Author: | [Not Specified] |
| Posted: | 2016-04-19 04:30 |
Hi ,
We have amonitoring system usingZen Packs based onZenoss 5.0 and with the exact same OS and product package we get the above issues on one site, while on two other sites with the same OS/product package we don't get that issue.
After install the below scenarios occur:
"""
Both instances of MetricShipper fail the health check with store_answering.
Zenpython, zenhub, and Zope are all failing the health check with zproxy_answering.
"""
This error occurs before any ZenPacks are loaded / any devices/events are configured.
After Multiple restarts, OS re-installs, package re-installs the issue still exists.
The System Requirements are met and exceeded on this server.
Has anbody found out why this happens on some systems and not on others with the same setup/packages
An upgrade is not possible in the new few months and a lot of re-development and testing is needed, so any help on wht these errors occur on some installs would be great or if anybody found a workaround for them.
John
| Subject: | I was never able to resolve |
| Author: | [Not Specified] |
| Posted: | 2016-04-19 11:36 |
I was never able to resolve that issue on my system. I ended up scrapping it and installing a fresh 5.1 system, which seems to be much more stable. Good luck with your issue.
| Subject: | this was related to dns. |
| Author: | [Not Specified] |
| Posted: | 2017-01-16 05:50 |
this was related to dns.
the containers communication was delayed by ~20 seconds due to dns lookups.
when installing zenoss in some cases the dns server in containers /etc/resolv.conf gets set to google dns.
if the server hasd no access to the external internet this causes 20 second delays in communication between containers which causes the apps health checks to fail in control centre.
solution was to set the dns server to a local ip on the contro, centre blade (set DNS1=... in the ifcfg files and restart newtork, NetworkManager, docker & serviced)
| < |
Previous WMI Monitoring |
Next How could I Monitore Multiples Postgres databases on the same server? |
> |