1

Starting at approximately 6 PM CDT ( 11 PM UTC ) on October 27, 2022, we will perform maintenance on the Spiceworks website that will affect News & Insights, Community, and our IT Tools such as Cloud Help Desk and Cloud Inventory. We anticipate this downtime taking no more than three (3) hours, but we’re scheduling for four (4) to be on the safe side. Based on that, maintenance should end no later than 10 PM CDT ( 3 AM UTC, Oct 28 ).

We will upgrade our database service instances. We will put the Spiceworks Community in full maintenance mode during this maintenance. You will be unable to log in to News & Insights but still be able to read the content. The community will be completely inaccessible, though.

Regarding the Cloud Help Desk (CHD), logged-out users will be unable to authenticate, and new user registration will not be possible until the maintenance is concluded. Aside from the Knowledge Base being inaccessible during maintenance, logged-in users will not be affected. Regardless of logged-in state, no data will be lost, and inbound emails will continue to process. This means it will be possible to interact with your help desk via email during the maintenance period.

Keep an eye out here or on our Status Page to keep up-to-date with our progress or any changes.

We appreciate your patience and understanding.

Update(s)

50 Spice ups
2

WOOHOO MAINTENANCE!!!

3 Spice ups
3
  • Do you consider complementing the status web page by further web and Cloud services for monitoring and reporting?

Currently, the status web page only lists

  • Spiceworks website www.spiceworks.com ,

  • Spiceworks community ( https://community.spiceworks.com and all its integrated components and integrations),

  • Spiceworks Cloud Help Desk (CHD) (and all its integrated components and integrations),

  • Spiceworks Connectivity Dashboard (and all its integrated components and integrations), and

  • Spiceworks Inventory Online (and all its integrated components and integrations).

It currently sees no needs to report on status of

I don’t know if Spiceworks review section, how-to section, scripts library section and further other sections are separate components with separate status. If yes, it would make sense to me to have them report their availability too and not be limited to be components to several of the reported services. And instead of just reporting operational or failed, some standard response times and their variation over time might be interesting too. I run more often into timeouts of various kinds and nonsense error web pages on the community platform than encountering an unplanned outage. I received several OOPs failed service or Bad Gateway web pages of existing web pages of the community in the past hour, despite the status reporting operational in green! Another reload was either changing the false error message or showing the web page claimed failed!

I’ve added corresponding feature requests to get monitoring of those listed services added and included into the Spiceworks status web page. Feel free to vote up and comment such feature requests if you would like them included too. When going on each item of the list above, you’ll get the reference to the corresponding feature request.

4

Will this bring any new features…like 2FA maybe?

5 Spice ups
5

Thanks for the heads up. Hopefully I’ll be asleep whilst this is being done.

6

a63346a2-24c2-472e-b8c2-efeada20161b-qzPu3i5.png
Will you post a Spicerex version of the Chrome dino game to entertain us during the downtime?

7 Spice ups
7

Anxiety starts

8

Are you sure you don’t want to do this on a Friday? Every insert IT title here starts updates on a Friday. :slight_smile:

2 Spice ups
9

And I thought that this downtime (23 - 03 UTC, so probably 01-05 in the early morning of your Summer time) overlaps with your sleeping time. Or will you have a night shift tonight?

1 Spice up
10
  • Didn’t you realize that this downtime will finish at a Friday morning UTC?

And don’t forget that the IT professionals are running their heterogeneous IT infrastructure with their internal hardware clocks in UTC anyways, reserving local times and time zones only to the user interfaces. Only these old Windows shops not yet having migrated to Cloud are still sticking to hardware clocks of their IT infrastructure in local time and time zones with those time jumps between daylight saving time and standard time twice a year, as they didn’t realize that even their operating systems got fixed to support heterogeneous infrastructure and UTC hardware clocks.

1 Spice up
11

A joke is what that was. The joke being never do system updates or any updates for that matter on a Friday because something can go wrong and your weekend is then stripped away from you.

1 Spice up
12

I should be sleeping then but for the chance to play a Spicerex game i can make exceptions

2 Spice ups
13

Good luck with the maintenance. See you on the other side.

1 Spice up
14

Hi scheff1, great suggestions. Let me bring that up internally. I know some of these can be an all-or-nothing issue (i.e., certain things that go down mean the Community is down), but some of these may be good to add to the list.

haha, I see what you did there.

We debated on what day to do this and ultimately decided on a Thursday for a few reasons. First, usually by 6 PM CDT (11 PM UTC), the vast majority of our audience and members are not online across the board (WWW, Community, CHD, etc.). Granted, the weekend is even lower, but the upside is that Friday, all of our developers are around. Should it look like the maintenance went off without a hitch but then there was an issue that popped up afterward; we still have everyone around to quickly address it.

I am normally a fan of doing these on Monday, but we want to do our best not to interrupt the service when members are using it most, so Thursday ultimately felt like the best day for this.

haha, do you mean like: https://chromedino.com/

4 Spice ups
15

Sure. That’s why I used feature requests for the case that we don’t get it with the current scheduled maintenance. I don’t like such features forgotten. And that is the purpose of feature requests as far as I understood.

Can’t follow the relation with what is already present. If the Spiceworks website or the community website is failing, then probably other services will be affected too. Not sure if the status web page would become failing then too. Far more often, I see increasing response times and not failing services. These increasing response times may lead to various strange effects without the service reported currently going to fail but some feature failing. E.g. I recently replied a topic where the OP had issues with loading an inventory view of Spiceworks Inventory Online. While Spiceworks Inventory Online was still available, loading of the records of such view was (probably temporarily) unavailable. When I read the topic hours later, I had no issues and asked if the symptom reported by OP was persisting. Before the OP could reply to my question of clarification, he did reboot his client device, some cleanup of caches, tried several web browsers, and the symptom was gone. So root cause was not possible to identify and most likely it had to do with response times, not of Spiceworks Inventory Online itself but of some service or component it was using. So even if an event will be an all or nothing event and the status page becoming unavailable, we may later see which events happened and which services were affected. And you know better which services and components are used in the background which might be meaningful to add to the list. If these would become many, than collapsing and expansion are meaningful options for grouping and make it easier to drill down or up.

16

We had to extend the maintenance on the Spiceworks Community specifically because the team ran into a few complications. Due to that, Maintenance went until 12:30 AM CDT ( 5:30 AM UTC ) on Friday (today).

If you run into any issues or see something that was not occurring before this maintenance, please do not hesitate to bring it up here. I saw @scheff1 ​ created another topic and it sounds like he and some others are experiencing random performance issues that are resulting in a 502, 504, or other errors.

If you do run into these kinds of issues, please let us know what page(s) and or action(s) you were attempting to make when you ran into this.

4 Spice ups
17

You know…I totally ignored the post about maintenance coming up but had this vague recollection of seeing October 27th on the post, which I didn’t remember until I hit the maintenance page last night…lol…I’m the guy that doesn’t see the “Use other door” sign until I pull on the door with that sign on it…Seriously…really appreciate the platform and the devs who make it all happen.

3 Spice ups
18

Whoops, I meant to reply to this one simultaneously, as I replied to others earlier in the conversation. But I failed, so let me get this one before I forget again.

Last night’s maintenance itself was geared more toward doing some significant updates to help not just get our tech stack up-to-date but also give us more flexibility on how we do a few things (especially concerning some background jobs that cause some, shall we say, hiccups steadily on Mondays and Thursdays).

That said, I do know people have been wanting MFA or 2FA and there is a CHD Feature Request that is getting quite popular that you (anyone reading this who agrees) may want to vote up:

https://community.spiceworks.com/feature_request/show/Cloud%20Help%20Desk%20(CHD)/6677

It is, of course, an internal topic, too. But we do often look at our Feature Requests as we refine our roadmaps, especially if there one that is in the same vein (or specific area of a codebase) as something else we’ll be working on. For example, the recent update for the initial release of the email templates (for CHD) .

@anthonytate4

19

I restarted to use the community platform not before 7:30 UTC. I had such issues at least until 11:35 UTC. Now, it seems better.

The web page with other named error is obvious. For this error Failed to load subscription data. , it was my subscribed web page in logged in view. These 502 and 504 errors were happening on same web page and on other web pages. All three of these errors happened by using the key shortcut in the current web browser Firefox (version 106.0.2) for reload of single currently active and selected web page which was the my subscribed web page. Sometimes it happened when creating a new tab for the my subscribed web page. The 504 error happened also on other web pages. Doing a reload with 504 error could lead to repetition of 504 error or to change to 502 error or to load of intended web page, especially the * my feed web page * in logged in view. On other web pages, I only got the 504 error. These other web pages were the community start page , my spice level web page , the daily challenge dashboard web page , the web pages 43 and 44 of the all time view of daily challenge leader board , the getting started field guide web page , the incoming activity web page , the inbox web page , and on almost any topic web page, regardless if of this year, 2019 or 2015. A click on a topic was sufficient, regardless if coming of a reference in an opened topic, the my feed web page or the subscribed web page. And the unanswered questions of general support forum web page gets the application error web page Oops! A Server Error Has Occurred. . Another symptom was that my avatar was missing most of the time and replaces by standard gray avatar of not having an avatar with no initials as if I would be in logged out view despite displaying subscribed web page or other profile web pages in version logged in view. When a topic with several replies was open, avatars were displayed for some random replies and standard gray avatar for other random replies. After 13:30 UTC, the platform seems to have recovered, don’t know for how long already as I was busy with other things.

20

I went to the experts and @erik-m-spiceworks ​ does a much better job of explaining this than I could:

It was time for a major version upgrade in the database engine of several related applications. At the end of the maintenance window traffic was reenabled but the application fell over immediately. This appeared to be a classic “thundering herd” problem but doubling application capacity wasn’t enough to keep the application standing and only spiked the database’s connection count and CPU. Code was rolled back to a week-old release to deal with some search-related exceptions that were ultimately unrelated. A few rounds of tweaking capacity and health check settings drew out the process, and very visibly since herds can only thunder outside of maintenance mode. In the end the database engine needed some additional cleanup (vacuum, recompute statistics, etc), some via manual intervention and some with the passage of time.

1 Spice up