This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Terrible support, terrible product.

I'm sure this post will be deleted long before it influences any potential customers. Too bad. For the lucky few who read this as you consider a purchase of Solarwinds products, you are hereby warned.

Imagine a support organization that answers open tickets infrequently, and when they do, it's after hours. Imagine dealing with squirrely software that requires constant patching, and that the patches constantly induce unstable behavior in other modules. Imagine spending money on a product that requires a HUGE investment of time and energy to maintain.

We've been using Orion NPM since November. It has been unstable for months. Here's the support routine:

Open a ticket.

Wait until after 5:00 pm to get an e-mail response asking for a 200 Mb diagnostic.

Upload diagnostic the next day. Wait until after business hours for a response.

Apply suggested solution. It doesn't work. Respond.  Wait until after business hours for a response. Nothing back. Give up. reboot the server. Problem temporarily goes away.

The pinheads who run support will no doubt take issue with the specifics. An e-mail wasn't received, Leapfile somehow lost the upload, but the point is true - this is not a stable product, and support is not worth waiting for much less paying for. Run now and select something else for network monitoring.

  • Dear Spudnet,

    I truly sympathize with your recent pains. I have had some negative experiences so I can relate. I can honestly  summarize our overall experience with the products and support to be more positive and successful. Hindsight is 20-20. Many of my own Orion issues/grief could have been avoided if:

    1) I made sure that all servers in my orion architecture had adequate resources - I monitored/alerted on orion services on all servers so that when something goes south - you find out fast.

    2) distributed across multiple servers (ie: multiple pollers, separate web consoles, honking sql) before issues arise.

    3) don't have too many administrators

    4) VM's can be your friend. Use "snapshot" before patching/upgrading. (don't use single cpu vm's for anything, min. dual cpu). Don't let VM admins tell you otherwise - save yourself some pain and demand it.

    5) Avoid using brand new releases of ANY BROWSER - these cause GRIEF to Orion admins.

    6) Ensure whoever is your Orion administrator is a cautious individual who employs self discipline, records changes and employs best practise techniques (ie: one change at a time, only touch 1 server at a time etc) when managing change. DO NOT LEAVE YOUR CORP ORION IN THE HANDS OF COWBOYS/COWGIRLS - no amount of support from any vendor can prevent outages in this scenario.

    Having been a previous Openview customer - I can say cost of ownership is less with solarwinds family of products. I am firm believer you get what you pay for. Solarwinds is economically a good deal. I want them to keep costs reasonable. I'm getting used to the overseas, customer support - recent experiences I've seen some great talent (not so much on the language skills), but great when they webex in and take control.

  • Got your e-mail, called and left you a message.

  • Epenney - Those are all good suggestions, and we actually follow almost all of them almost all of the time. I think we're pretty well provisioned. Migrating to Solarwinds was a big deal and we put the resources behind it.

    I agree with your "honking SQL" comment. We had a lot of trouble after we moved from the test phase and SQL Express. We had the DBAs do some tuning and things settled down. The most concerning issue is that interface stats just stop accumulating without so much as a blip. I report on various metrics and feel like an idiot explaining big holes in the charts.

    We saw a lot of favorable comments on thwack prior to purchase and judged that support was available and responsive. Solarwinds' initial support in November was a LOT better than is has been over the last 30-60 days. Our experience has been that responsiveness has deteriorated at the same time that our issues have increased. Not too surprising, but unfortunate nonetheless.

    We have had persistent issues beginning in February, but no tech has suggested a remote session to dig in further. I understand from other responses that they have had better experiences. I expected this same level of support but have not seen it.

    I put a call in to Brandon, but haven't heard back yet. I'll update the post as we work through this.

  • Just to correct an incorrect/inaccurate statement on this thread.  Our support is not out sourced, it is staffed by SolarWinds employees.  To provide 24x7 coverage across the world and all our customers we have SolarWinds offices for Support here in Austin in the States, Cork, Ireland and Singapore.  These are not out sourced call centers or anything like that.  We load balance this load to try to get back to customers as quickly as possible.  Sometimes that means you may have a rep from one of our other offices assisting you.

  • I recently had same "red faced" experience of having blank stats for interfaces.

    2 separate "red faced" scenarios happened to me.

    1st situation - numerous blank gaps on various interfaces, random devices, no particular pattern. this particular Polling Engine is a VM. Turns out the VM only had single CPU, 3gb ram. Since adding 2nd cpu & increased ram to 4gb VM - blank stats in interfaces for the most part vanished.

    2nd scenario - was far worse - db sync'ing for entire poller would stop without warning us (result no data collection for interfaces, but oddly nodes still reporting up, response time/packet loss still being captured. For this I created an alert. I refer to it as "Canary Node Alert". For each polling engine, I select a major node that is being polled by that engine (one that is rarely offline/down). Alert checks LastSync value to see if its gt 5 minutes - if so it rings alarm bells. There is probably a better way to alert on polling engine db sync health - in my urgency to ensure this never repeated I created the alert described above. Attaching my "canary" alert - i cloned this alert for each polling engine we have.

     

     

    I really hope my "red faced" days are behind me.

    Poller-Syncronization-Alert.zip
  • Spudnet,

    I apologize for any inconvenience we might have caused you. I lead the support organization at Solarwinds and would like to be of assistance with any issues you are experiencing. Please reach out to me directly at olu.aluko@Solarwinds.com

     

  • I doubt you'll get a response, thread was necro'd, originally posted in 2011. Lets hope he got his issues fixed though!