I would like to do gracefull exit to my nodes.
i have 3 of them.
i gave the command in my synology
storagenode exit-satellite and after i gave one by one all satellite domain names
i checked and the command was delivered…(by trying the same procedure again it didnt find any sattelite available to exit)
it is a month after and every time i try to check the status with the command
docker exec -it storagenode /app/storagenode exit-status --config-dir /app/config --identity-dir /app/identity
i see the progression is at 0%
I think you should check the logs, because when the command is given the screen will not update. The terminal screen is not captured on the progress basically.
You need to search for grace in your logs.
However, if the command has been succeed, the final state would be either 100% and Y or 0% and disqualified on your dashboard after the end of the GE.
When you remove the container, the previous logs will be gone, so if you started more than a two weeks ago, likely the logs were cleaned up already.
Then you need just wait until it would finish.
i gave up. I waited 37 days… I shut down my servers and my nodes without graceful exit. I can’t understand what went wrong.
Generally i had many problems with my nodes under docker in my synology nas. Now i created new storagenodes in windows environment, with the Win GUI Storj Node Toolbox given.
Since now (2 days running) no problem at all. No stops , no dropdowns without reason…easy installation for multiple nodes. I had many problems with docker and had to restart the nas and the nodes 2-3 times per week. In windows 11, everything looks so steady…i also created a task to check my node status and if down to restart the server automatically, so i don’t bother to controll all the day as i did when newbie
Interestingly, I went opposite direction. I shut down Windows node and turned on Synology node.
The reason - as discussed here, after the node size went over some 12 TB, on NTFS with all possible improvement settings, the filewalker becomes nevereding story - never completes its job.
But you may be more lucky
My nodes are up to 12GB. i have 3 of them. I run windows 11 latest version and i avoid the storj installation completely. I found the Win GUI Storj Node Toolbox much faster and better. I would say more stable. My nodes are up for a week and everything works fine. I am in vet period ofcourse so my traffic is still low. But non stop running of all of my nodes for more than 7 days was something very rare in synology docker…they where drop down without a reason probably from irregular storj pieces in download or in upload.
I think when you signal GE you’re essentially just telling the Satellite “I plan to leave: if you need any data from me ask for it soon”. Then the repair system can look at all the data you’re holding pieces for… and decide if it should repair anything to get those piece counts higher.
If that’s how it works… then if there are a healthy number of pieces for everything a node is holding… then GE does nothing but prevent new files from being sent to that node. It may not need to download/repair anything at all - so it may transfer nothing for those 30 days.
That’s just what I’ve gleaned for other posts: someone is sure to correct me if I’m wrong. But it would mean you can’t really speed up GE: as in healthy scenarios it may not do anything other than wait those 30 days (with the node being a priority source, if repairs are needed, so you qualify to have your withheld coins returned)
Could be…altought (if that´s the case) I´d argue it made more sense to just delete the pieces “not needed” from my node, assuming a Healthy scenario.
Well, good thing is that Holding is just 17$
Too much work for the satellites, this is why we changed the behavior to described by @Roxor
You just need to keep it online with an online score not less than 80% for the next 30 days after you call the GE.
None of the options to speedup, because pieces are not transferred from your node by your node to other nodes in the network as was before. But to prepare the list of those nodes satellites spent too much resources and time, also you may have a bad connection, which would prevent your node from transfer data without failures and thus risk of disqualification was much higher.
