Back revving Chia version

Some months ago, I upgraded from version 1.1.7 to 1.2.9.
Ever since then, I have had problems with syncing and not having all of my plots recognized. Right now, I have been out of sync for several days, and keep falling further behind.

I never had any issues with 1.1.7
But I have no idea if that version would be problematic today, or if back revving is as easy as forward revving?

Has anyone ever back-revved to an older Chia version?

Or should I install the latest version?

I have not been keeping abreast of which versions have which problems.
Is there a version noted for its stability and absence of syncing issues?

Thank you.

I’m on .2.11 there is no point running any versions before that, as if you drop sync for more than 20 blocks you will never resync ( that’s advice from chia devs ).

Do I still lose sync on .11, yes, but its now pointless running an earlier version due to changes.

2 Likes

That advice was only with respect to v1.2.9, and it looks like has been proven wrong. Also, even with that advice, that version is not blocked for downloads.

Actually, I was running v1.2.6 till about the end of December. At that time I decided to upgrade to v1.2.11, but for whatever reason I couldn’t do it. I also found posts from other people trying to upgrade from v1.2.6/7 directly to .2.11 having the same problems. The only option was to upgrade to v1.2.10 first, let it stabilize, and then upgrade to v1.2.11

There are some log entries issues with both .2.6 and .2.7 during dust storms, so I would suggest not going backward, but rather upgrade to .2.11 directly, or if that doesn’t work, upgrade to .2.10 first and once that starts, upgrade to .2.11.

That’s not what I read, I read any earlier version than .11

What was the issue with trying to skip versions and go directly to 1.2.11?
I am asking in order to make an informed decision on whether to do so, or to first install 1.2.10?

What issues did you have and others have?
Also, if I run into a problem going straight to 1.2.11, then will it still be a simple matter to install 1.2.10?

There is a thread about it on github either Issues or Discussion pages, so you can search for it there.

v1.2.11 just refused to run. Kind of looking back, maybe that was something around ca folder, but I am not sure. Installing v1.2.10 was as easy as it could be. After that v1.2.11 also run after one install. So, clearly, v1.2.10 cleaned up something for v1.2.11 to run.

It may be that the problem is there only for those installing from v1.2.6/7 directly to v1.2.11. I cannot tell which version is crucial to have in between those, whether you will be good with v1.2.9 to do it in one step. Although, the two steps upgrade is also painless and fast.

1 Like

Take a look at this posts (third bullet). He later clarified it further down to those 20 blocks.

Maybe some other person posted that stating that the problem is for all versions.

2 Likes

To be fair, I dont recall where I read it.

1 Like

I just installed 1.2.11
It was fast and painless.

Syncing is working, but I am approximately 5,000 units behind being fully synced.

Any idea how long it will take to be back in business?

1 Like

Please upgrade to 1.2.11

Done.
Any idea how long it will take for 5,000 units behind to catch up?

That depends mostly on your H/W (CPU). But, it should not take too long (long minutes, I guess, not hours). Just monitor your cores, not the overall CPU. (If you don’t see your cores choking, you can try to lower the peer count.)

I need a couple of beers first…

I am synced!

And my previously not recognized plots (1,500 plots) are now all registered.

2 Likes