The argument for official support vs third party support
-
@scottalanmiller said in The argument for official support vs third party support:
@black3dynamite said in The argument for official support vs third party support:
@wrx7m said in The argument for official support vs third party support:
Obviously, there is value to knowing why something occurred. For instance, when it occurs repeatedly. Then you have to identify the root cause.
When I have an extra desktop or laptop ready for deployment. I usually just swap the machine and then take my time to identify the issue.
Same with a hard drive. In a lot of cases, having parts available is far more reliable, faster, and cheaper than a warranty from a vendor.
True story. I just bought a 12-bay Synology and 14 identical drives.
-
@wrx7m said in The argument for official support vs third party support:
@scottalanmiller said in The argument for official support vs third party support:
@black3dynamite said in The argument for official support vs third party support:
@wrx7m said in The argument for official support vs third party support:
Obviously, there is value to knowing why something occurred. For instance, when it occurs repeatedly. Then you have to identify the root cause.
When I have an extra desktop or laptop ready for deployment. I usually just swap the machine and then take my time to identify the issue.
Same with a hard drive. In a lot of cases, having parts available is far more reliable, faster, and cheaper than a warranty from a vendor.
True story. I just bought a 12-bay Synology and 14 identical drives.
I lean more and more to this all of the time. MOre often than not, vendors fail to deliver on their support promises anyway.
-
@scottalanmiller said in The argument for official support vs third party support:
@wrx7m said in The argument for official support vs third party support:
@scottalanmiller said in The argument for official support vs third party support:
@wrx7m said in The argument for official support vs third party support:
@scottalanmiller It is the value of time spent troubleshooting vs time spent replacing. Sometimes it makes more sense to replace something than try and figure out what it is. If I spend 2 hours trying to figure something out on a $100 printer, I should have just replaced it.
Yup, which is often the case with operating systems, too!
Exactly. A re-image solves almost everything.
Yup, the DevOps philosphy. Rebuilt to know good rather than trying to troubleshoot the unknown.
This is fun until it involves hardware. Ever try to downgrade a firmware on a NIC or SSD? For some vendors this is technically impossible without shipping it back....
-
@scottalanmiller said in The argument for official support vs third party support:
Same with a hard drive. In a lot of cases, having parts available is far more reliable, faster, and cheaper than a warranty from a vendor.
The quality of firmware and drivers on flash devices is a lot more all over the place than magnetic drives. Having 14 drives that implode under a burst of writes isn't that helpful.
-
@scottalanmiller said in The argument for official support vs third party support:
Commodity parts, get one that works and move on. Paying for unique support and code changes to support a single SMB use case rarely works. As Dustin points out, I can swap the part in an hour. I can't get support to understand what the issue is in that amount of time. Vendor support, in that example, represents high cost and high risk. I know that I can swap parts and that I can swap them now. I don't know if the vendor can fix the issue or if they will in any reasonable amount of time.
This only works if you keep non-matching spares. So For every LSI based controller you keep a similar Adaptec (and are ready to do a swing migration). For every Intel NIC have a QLogic, for every Intel Flash drive have a HGST one.
I watched (larger server OEM) techs replace a back plane twice and following that we got routed to the right team who got us a hotpatch for a SAS expander.
What's more fun with server OEM's is while the back end vendor may have identified the issue (say Broadcom), they run different code trains, and they may not check out that fix from the main OEM's trunk (for fear it causes other issues) until a customer reports an issue. Because of the hell of properly regression testing hardware this is how a lot of stuff works.
-
@storageninja said in The argument for official support vs third party support:
@scottalanmiller said in The argument for official support vs third party support:
Commodity parts, get one that works and move on. Paying for unique support and code changes to support a single SMB use case rarely works. As Dustin points out, I can swap the part in an hour. I can't get support to understand what the issue is in that amount of time. Vendor support, in that example, represents high cost and high risk. I know that I can swap parts and that I can swap them now. I don't know if the vendor can fix the issue or if they will in any reasonable amount of time.
This only works if you keep non-matching spares. So For every LSI based controller you keep a similar Adaptec (and are ready to do a swing migration). For every Intel NIC have a QLogic, for every Intel Flash drive have a HGST one.
I watched (larger server OEM) techs replace a back plane twice and following that we got routed to the right team who got us a hotpatch for a SAS expander.
What's more fun with server OEM's is while the back end vendor may have identified the issue (say Broadcom), they run different code trains, and they may not check out that fix from the main OEM's trunk (for fear it causes other issues) until a customer reports an issue. Because of the hell of properly regression testing hardware this is how a lot of stuff works.
Same problem for the vendors. If you are dealing with unpatched spares, so are they. Having worked for some of the big ones, I know that their supply chains struggle to get parts, too. Heck, IBM couldn't deliver a server internally in more than six months, imagine how hard it is to get support parts!
-
This post is deleted! -
@scottalanmiller said in The argument for official support vs third party support:
Same problem for the vendors. If you are dealing with unpatched spares, so are they. Having worked for some of the big ones, I know that their supply chains struggle to get parts, too. Heck, IBM couldn't deliver a server internally in more than six months, imagine how hard it is to get support parts!
Unpatched spares are supposed to be handled by the support staff, but that gets overlooked way too often.
We've actually been building life cycle tools into the hypervisor to mitigate this (because expecting SMB's to deploy HP One View etc, is a non-starter. ESXi can patch RAID controllers, and SAS HBA's today with more coming in the future.
-
@scottalanmiller said in The argument for official support vs third party support:
@storageninja said in The argument for official support vs third party support:
@scottalanmiller said in The argument for official support vs third party support:
Commodity parts, get one that works and move on. Paying for unique support and code changes to support a single SMB use case rarely works. As Dustin points out, I can swap the part in an hour. I can't get support to understand what the issue is in that amount of time. Vendor support, in that example, represents high cost and high risk. I know that I can swap parts and that I can swap them now. I don't know if the vendor can fix the issue or if they will in any reasonable amount of time.
This only works if you keep non-matching spares. So For every LSI based controller you keep a similar Adaptec (and are ready to do a swing migration). For every Intel NIC have a QLogic, for every Intel Flash drive have a HGST one.
I watched (larger server OEM) techs replace a back plane twice and following that we got routed to the right team who got us a hotpatch for a SAS expander.
What's more fun with server OEM's is while the back end vendor may have identified the issue (say Broadcom), they run different code trains, and they may not check out that fix from the main OEM's trunk (for fear it causes other issues) until a customer reports an issue. Because of the hell of properly regression testing hardware this is how a lot of stuff works.
Same problem for the vendors. If you are dealing with unpatched spares, so are they. Having worked for some of the big ones, I know that their supply chains struggle to get parts, too. Heck, IBM couldn't deliver a server internally in more than six months, imagine how hard it is to get support parts!
Shit like this just blows my mind.
-
@dashrender said in The argument for official support vs third party support:
Same problem for the vendors. If you are dealing with unpatched spares, so are they. Having worked for some of the big ones, I know that their supply chains struggle to get parts, too. Heck, IBM couldn't deliver a server internally in more than six months, imagine how hard it is to get support parts!
Shit like this just blows my mind.
Parts Bins, internal supplies for labs, and customer supply chains are all completely different (well IBM may have been a gong show). Dell and HPE staff can't just go grab something off the line, with Mfg you have to account for the costs and someone gets to pay (and often at a premium to prevent abuse) for those internal servers.
Parts Bins and stocking those are different, and supply chain for a OEM might actually be different in the us than EMEA.
-
@storageninja said in The argument for official support vs third party support:
@dashrender said in The argument for official support vs third party support:
Same problem for the vendors. If you are dealing with unpatched spares, so are they. Having worked for some of the big ones, I know that their supply chains struggle to get parts, too. Heck, IBM couldn't deliver a server internally in more than six months, imagine how hard it is to get support parts!
Shit like this just blows my mind.
Parts Bins, internal supplies for labs, and customer supply chains are all completely different (well IBM may have been a gong show). Dell and HPE staff can't just go grab something off the line, with Mfg you have to account for the costs and someone gets to pay (and often at a premium to prevent abuse) for those internal servers.
Parts Bins and stocking those are different, and supply chain for a OEM might actually be different in the us than EMEA.
At IBM< we were an external customer, even though we were inside IBM. We showed up just like any external enterprise customer. So their inability to support was universal.
