> I don't know this for sure, but from my own upgrades and user feedback I'm
> afraid that the "major version" of the state file changed somewhere between
> 4.14 and 4.17, so an update from 4.17 (and possibly 4.16) to 4.19 should work
> without problems, but upgrading from prior versions might fail with this
> error. I didn't look into the the code, though, anyone else to confirm this?
>
> BM
>
But an upgrade from 4.15 to 4.17 is also not possible on my box. Can it be usefull to deattach Einstein from this computer (4.15), then re-attach and then upgrade?
> But an upgrade from 4.15 to 4.17 is also not possible on my box. Can it be
> usefull to deattach Einstein from this computer (4.15), then re-attach and
> then upgrade?
Glaub ich nicht. Ich wuerde, wenn mir die CPDN so wichtig ist, die CPDN WU mit 4.15 zuende rechnen, dann alle Pojekte "Updaten" und dann das Upgrade auf 4.19 machen und den Reset der Projete dann riskieren.
> I don't know this for sure, but from my own upgrades and user feedback I'm
> afraid that the "major version" of the state file changed somewhere between
> 4.14 and 4.17, so an update from 4.17 (and possibly 4.16) to 4.19 should work
> without problems, but upgrading from prior versions might fail with this
> error. I didn't look into the the code, though, anyone else to confirm this?
>
> BM
The version number used in the state file is the version number of the client.
For the clients that have experienced a complete reset, it generally means that the new client believes that the state file has been corrupted.
> I don't know this for sure, but from my own upgrades and user feedback I'm
> afraid that the "major version" of the state file changed somewhere between
> 4.14 and 4.17, so an update from 4.17 (and possibly 4.16) to 4.19 should work
> without problems, but upgrading from prior versions might fail with this
> error. I didn't look into the the code, though, anyone else to confirm this?
I noticed it when going from 4.13 to 4.15. All the other releases (.16, .17, .18, .19) haven't had any issues. I never tried 4.14, so I can't comment on whether the change was implemented there or not.
My upgrade to 4.15 didn't affect all projects though, only Predictor and E@H. CPDN and S@H were unaffected. It didn't actually say it was resetting the projects, it simply wiped out the WUs on those two projects.
> I don't know this for sure,
)
> I don't know this for sure, but from my own upgrades and user feedback I'm
> afraid that the "major version" of the state file changed somewhere between
> 4.14 and 4.17, so an update from 4.17 (and possibly 4.16) to 4.19 should work
> without problems, but upgrading from prior versions might fail with this
> error. I didn't look into the the code, though, anyone else to confirm this?
>
> BM
>
But an upgrade from 4.15 to 4.17 is also not possible on my box. Can it be usefull to deattach Einstein from this computer (4.15), then re-attach and then upgrade?
wolmic
> But an upgrade from 4.15 to
)
> But an upgrade from 4.15 to 4.17 is also not possible on my box. Can it be
> usefull to deattach Einstein from this computer (4.15), then re-attach and
> then upgrade?
Glaub ich nicht. Ich wuerde, wenn mir die CPDN so wichtig ist, die CPDN WU mit 4.15 zuende rechnen, dann alle Pojekte "Updaten" und dann das Upgrade auf 4.19 machen und den Reset der Projete dann riskieren.
BM
BM
> I don't know this for sure,
)
> I don't know this for sure, but from my own upgrades and user feedback I'm
> afraid that the "major version" of the state file changed somewhere between
> 4.14 and 4.17, so an update from 4.17 (and possibly 4.16) to 4.19 should work
> without problems, but upgrading from prior versions might fail with this
> error. I didn't look into the the code, though, anyone else to confirm this?
>
> BM
The version number used in the state file is the version number of the client.
For the clients that have experienced a complete reset, it generally means that the new client believes that the state file has been corrupted.
----- Rom
----- Rom
My Blog
> I don't know this for sure,
)
> I don't know this for sure, but from my own upgrades and user feedback I'm
> afraid that the "major version" of the state file changed somewhere between
> 4.14 and 4.17, so an update from 4.17 (and possibly 4.16) to 4.19 should work
> without problems, but upgrading from prior versions might fail with this
> error. I didn't look into the the code, though, anyone else to confirm this?
I noticed it when going from 4.13 to 4.15. All the other releases (.16, .17, .18, .19) haven't had any issues. I never tried 4.14, so I can't comment on whether the change was implemented there or not.
My upgrade to 4.15 didn't affect all projects though, only Predictor and E@H. CPDN and S@H were unaffected. It didn't actually say it was resetting the projects, it simply wiped out the WUs on those two projects.
It did generate new host IDs.
OK, all 6 machines running on
)
OK, all 6 machines running on 4.19 without incident. Two are Win2K, the rest are XPSP2. No lost WU's so far, no crashes or freezes.
Happy crunching!