Parity taking days to (fail to) sync on warp-barrier #8531
Comments
|
Having deleted the data dir and restarted with only |
Tbaut
added
F2-bug 🐞
|
Same here. Seems to start with warp sync, then switches to full sync.
My system specs are as follows: CPU: Intel® Core™ i7-7500U CPU @ 2.70GHz × 4 Output: |
|
CTRL+C'ing then rerunning with the same command appears to be working now. At snapshot 365 of 1337 at present. Will update if it completes or switches back to full syncing again. |
|
Just got a full sync. Total time about 2.5 hours. @cachvico - Instead of deleting |
|
I have restarted first with |
|
@cachvico - Does it get further for you now after the restart? Mine didn't complete the first time, but completed on the first restart. |
|
@JSterling8 not really - still only up to #4638589 I've updated my log at #8531 (comment) |
|
Great! Which branch should I build from to pick up the fix? (Or, when's the next release that includes it?) Thanks |
|
@cachvico You can try |
|
And we're off! Fingers crossed
|
|
I deleted Observe that I still seemed to need to hit Ctrl-C and restart, in order for sync from snapshots to work: |
|
In fact it seems to choose randomly between snapshot sync and regular sync when restarting after ctrl-c |
|
@cachvico Actually the logs saying "Syncing #0" is quite misleading because it could be that the node is just waiting for valid peers to start the snapshot. So the issue here is that it can't find one.
Just to be sure, if you could provide |
|
(The issue on the mainnet is that currently there seem to be only one peer serving snapshots) |
|
Obviously the |
|
Yeah sure, just the first minute should be fine. |
|
Here we go: $ ./target/release/parity --db-compaction ssd --cache-size 1024 --warp-barrier 5520000 -l sync,warp,snapshot,network=trace > ../output.txt 2>&1` |
|
So reading from the logs, it seems that the only peer that could provide a snapshot ( |
|
I deleted my network folder and restarted with I've been trying to sync with Parity for 10 days now.. should I restart without |
|
Restarted without warp, in branch \o/ |
|
Daily update : we actually found 2 bugs you might have run into:
With those 2 fixes you should be able to sync in 1h30 now. |
|
The 2nd one should not be on 1.10 branch. |
|
Closing this issue as the latest PRs should have addressed it. Reopen if it persists. |
|
I've been using Parity since my successful sync two weeks ago, but didn't have too much luck on a clean restart today. Can I confirm the fixes are merged into Thanks a lot |
|
@cachvico The first two times you're shutting your node down while it's trying to download a snaphot? And it's actually successfully picking up where it left off. The last time I think you're being hit by a bug that's only on master that's a deadlock issue, it's being fixed right now (in general building and running from master is not always the most reliable). |
|
Thanks @folsen - which is the branch to build from to pick up all the fixes referenced in this issue? |
|
@cachvico I don't know, it depends on what you're trying to do, are you trying to develop something and work on a fix for something? Otherwise I'd just recommend to download the latest release. |
|
I'm simply trying to run the Parity client (e.g. for blockchain querying via web3). Sorry can you confirm which release version to use? My build is on 1.12.0-unstable, but the latest release on the website is 1.10.4. Given all the problems I had earlier (hence this Issue thread), I don't want to try again with the wrong version.. |
|
Try 1.11.1 from here https://github.com/paritytech/parity/releases |
|
Built 1.11.1 from source and it's running, however not sure it's working. Output attached - first run from clean build after 2 hours. (update: 8 hours and still no apparent progress; ~/.local dir is 2.4M) Command used was Couple of more minor observations:
Many thanks for all the follow-ups. |
|
@cachvico thanks for your feedback. A newcomer guide isn't available right now but a work to make our FAQ more beginner friendly has already started. For such cases as yours, please join our Gitter channel to get to understand how things work under the hood, not only from the parity team but also from the community and more importantly help you resolve your problems. |
|
Sorry what does "cases like mine" mean? Please note that since opening this Github Issue (at your request @Tbaut on StackOverflow) I still don't have a functioning Parity client despite following instructions and due diligence with clean builds, etc. Ever so slightly frustrating to be given the brush-off here. |
|
@cachvico A case like yours would be where someone despite (your) best efforts can't get parity running properly, we've had this back-and-forth in this issue for over a month. Most people don't have this experience, and having one message back and forth per day for another month is not the most effective way to deal with it. If you jump into chat with us, we can probably get it resolved faster. Edit: Also, the original thing that @Tbaut asked you to raise a bug about in the StackExchange thread was that |
|
To be clear, I was running the fix set up for me by yourselves, for 2 weeks of that month. You then (reasonably) closed the bug but I was stranded on a dev branch so now all I need to do is build from the fixed release. Except the fixed release given is not working, so back to square 1 (in fact a step backward because now I don't have a functioning node at all, which I did before). I can now be reached on the Gitter channel and will post a message in due course.. |
|
@cachvico Yeah I feel you, you've been hit by a cacophony of different issues, all of them should have been fixed, but I think you may just need some patience until we cut a 1.11.2 release to get all of them. Building from master you hit another issue and we don't recommend running a node that is built from master. You'd probably have more luck going back to 1.9.x or 1.10.x or some stable release than trying to rely on beta/alpha builds. 1.11.1 should work for you, but for some reason your sync got stuck, there can be a ton of reasons for this, thus it's easier to debug on chat. Hopefully it'll be resolved shortly. |
|
Tbaut verified my above results in chat, confirmed fast sync from clean start is currently unavailable, and advised me to keep retrying. I'm trying |
|
@cachvico It's related to #8710 to be patched in 1.11.2 |
|
OK. I'm back on geth in the meantime, completed a full sync in about two days. |
I'm running:
Parity taking days to (fail to) sync despite multiple restarts as follows:
Initial run started with canonical
parity- cancelled after a couple of days, deleted~/.local/share/io.parity.ethereumparity --mode active --tracing off --pruning fast --db-compaction ssd --cache-size 1024- again cancelled, deleted parity dir and then:parity --db-compaction ssd --cache-size 1024 --warp-barrier 5520000This last execution was left running for 48 hours and the last line of the log at that point:
2018-05-02 15:52:13 Syncing #4179651 a017…7492 5 blk/s 434 tx/s 28 Mgas/s 0+ 453 Qed #4180107 25/25 peers 66 MiB chain 393 MiB db 43 MiB queue 13 MiB sync RPC: 0 conn, 0 req/s, 44057 µsMy system specs are as follows:
CPU: Intel i7 6700K
GPU: GeForce GTX 1060 6GB
RAM: 16GB DDR4 SDRAM
HD: 128GB M.2 SSD
Internet: 950 Mb/s down, 120 Mb/s up, fiber
Clock is synced. 8GB swapfile is 152M used.
StackExchange discussion here, https://ethereum.stackexchange.com/questions/46950/parity-sync-time-april-2018-many-days-normal
Raising a bug as requested by Tbaut. I would be happy to help with diagnosis and testing.
Thank you,
Darren
The text was updated successfully, but these errors were encountered: