Why On Premises Spiceworks Is Impacted
-
We are getting a lot of customer questions and concerns around failures of the on premises Spiceworks installations due to the ISP outages for the Spiceworks community. While the information about why this is happening is public, many people using on premises installs have not looked into this and are unaware of why there is impact.
On premises Spiceworks is still very heavily tied to the cloud components, especially to the community. The authentication mechanism of Spiceworks is hosted, not local, and so authentication calls need to be able to get to Spiceworks Community to complete.
Important note: If you manage to get logged into your SW install, do not log out, keep the session going. You should be okay as long as you remain logged in. If you do get logged out, just keep retrying, the community is up and down and you only need to authenticate once to get back in.
Some portions of the on premises system are actually in the cloud, such as the purchasing component, so these portions will not be expected to work. But the on premises scanning and helpdesk portions should continue to work as long as authentication has not been needed.
-
Spiceworks is working to modify things to get on premises working again:
-
Question is how are they going to get the patch out if they're always down?
-
@NerdyDad said in Why On Premises Spiceworks Is Impacted:
Question is how are they going to get the patch out if they're always down?
Patching should be able to be moved to other hosted independent of other systems, I would imagine. It's just file hosting, not a massive database that has to be kept in sync.
-
Understandable. That just may have to be something that they port by sneakernet to the outside world.
-
@NerdyDad said in Why On Premises Spiceworks Is Impacted:
Understandable. That just may have to be something that they port by sneakernet to the outside world.
Their offices are online, just their hosting is down.
-
I wonder how long the advertising company took before they realized that their hosted solution was having a problem?
-
@Dashrender said in Why On Premises Spiceworks Is Impacted:
I wonder how long the advertising company took before they realized that their hosted solution was having a problem?
The advertising company? Who do you mean?
-
@scottalanmiller said in Why On Premises Spiceworks Is Impacted:
@Dashrender said in Why On Premises Spiceworks Is Impacted:
I wonder how long the advertising company took before they realized that their hosted solution was having a problem?
The advertising company? Who do you mean?
Damnit - two times in an hour... I meant marketing company.
-
@Dashrender said in Why On Premises Spiceworks Is Impacted:
@scottalanmiller said in Why On Premises Spiceworks Is Impacted:
@Dashrender said in Why On Premises Spiceworks Is Impacted:
I wonder how long the advertising company took before they realized that their hosted solution was having a problem?
The advertising company? Who do you mean?
Damnit - two times in an hour... I meant marketing company.
Do you mean how long before the people in the office realized that the site was down? Pretty quickly, I'd imagine, as they would have alerts, and the mod and publishing staff use it constantly.
-
Since they are external users to the site, they would have seen it go down just like we did.
-
Otherwise, people like @SeanD would have been able to have been posting updates constantly that would have "appeared" as soon as the site was up, even if only for a minute. But he's stuck posting only when the site is up just like us. So they are using an external link to the site, not some back channel private line to the datacenter. Which makes sense, they get full visibility into what is going on for end users that way. Same as on ML.
-
My comment was meant as both a joke and social commentary.
-
So wait, so the on prem installation ties to the cloud. So the cloud is a dependency. But the cloud is down because there is only one ISP and no failover? So basically, no matter how much RAID, no many servers, how many backups, how many failover ISPs the customer has, no matter how expensive the routers that they use and redundancy switches.... one single ISP goes down and every single installation in the world goes offline at the same time because they are all dependent on the same single point of failure? No redundant ISP, no redundant datacenter? No failover plan at all?
What if that one datacenter was to burn down. Would there even be a way to bring the on prem systems back up, ever?
-
There is a rumour that on premises SW can be improved by blocking this URL: frontend.spiceworks.com
I am not testing this locally, so cannot speak to it working. Tagging @Coloradogeek who shared this with me. I'm not certain if this just speeds up the slow on prem issues, or if it somehow tackles the authentication breaks.
You would block this in the firewall to keep the on prem application from attempting to talk to the site.
-
@StrongBad said in Why On Premises Spiceworks Is Impacted:
So wait, so the on prem installation ties to the cloud. So the cloud is a dependency.
This is not news. they implemented this like 2 years ago now, maybe 3.
-
@scottalanmiller said in Why On Premises Spiceworks Is Impacted:
There is a rumour that on premises SW can be improved by blocking this URL: frontend.spiceworks.com
I am not testing this locally, so cannot speak to it working. Tagging @Coloradogeek who shared this with me. I'm not certain if this just speeds up the slow on prem issues, or if it somehow tackles the authentication breaks.
You would block this in the firewall to keep the on prem application from attempting to talk to the site.
Anyone get a chance to test this, yet? I know that things are mostly working at the moment, but was hoping for some feedback for future issues.
-
Another reason why I want to move helpdesk systems
We don't use the network scanner or inventory any more. -
@hobbit666 said in Why On Premises Spiceworks Is Impacted:
Another reason why I want to move helpdesk systems
We don't use the network scanner or inventory any more.We had customer after customer request migration off of SW yesterday. I'm not sure which factor pushed them to do so, but my guess is the fragility being exposed that it has all of the risks of on premises installs combined with all of the risks of hosted installs combined with not being redundant in location or ISP connections. Some that didn't believe or understand that the on prem was tied to the cloud might be upset about that as well, but they should have known and if that is their reaction they are clueless and the reaction is inappropriate.
-
@scottalanmiller To what platform are you moving them? Is there a good way to extract the existing data in SW?
