Self build EmbassyOne. Bitcoin Core intermittent and constantly timing out. Having to either restart or stop/start with veeery long restart periods. 150Mbs down | 150 Mbps up connection. Connected directly to router.
What are your hardware specs and how many services are you running?
I am having the exact same issue since the system update version 0.3.4~2. Sometimes bitcoin core has a yellow triangle and sometimes green dot. When it has a green dot, eletrs is spinning trying to sync. Its back and forth. Tried to restart and it never does. Just sits there at the “restart in progress” screen. Has this been answered or have you found a solution. The actual help on the issue from start9 is slim im noticing.
What are your hardware specs?
embassy one (canakit for umbrel) but installed embassy OS on SD card. 8G memory, 2TB drive. Been working great for 6-7 months.
Intel Pentium G3240 3.10Ghz 64 bit
Could the RAM on the Raspberry used in the self build be a contributing factor? I think it might be 2GB RAM?
You definitely need more than 2GB of RAM, and on a Pi, we highly recommend 8GB, as the 4GB model is known to have issues with Bitcoin.
I have the same issue after update to 0.3.4~2. Bitcoind timeout. Its on my Pi with 8gb of ram. I have no issues on the second one on x86. Same version of startOS.
Same hardware and I keep having the same issue as well. And only started after the update to 0.3.4~2
Seems to fix the issue when I hard restart (power off) the raspberry pi. But it inevitably acts up again.
If you go to:
Services > Bitcoin Core > Config > RPC Settings > Advanced
What are your settings for:
Known good settings are:
Work Queue 128
It could be that yours are stuck lower than that and causing health checks and dependent services to fail for bitcoin.
Also, take a look at your ram usage @ System > Monitor > Memory
What values are there for Total and Available?
If you have little ram left, try shutting down some services that you don’t actually use.
Last, if you’ve done all that and it looks good and yet the problems persist, it might be worth trying
System > System Rebuild.
My work queue was at 64. Trying 128.
Thanks very much for the input and recommendations George.
RPC settings for Threads and Work queue are the same as those you’ve indicated.
RAM - Total: 1899.01MiB ; Available: 1304.50MiB; Used: 500MiB -
Only service currently running is Bitcoin Core.
I’ve considered System Rebuild option but hesitant as I don’t know the consequences of doing so on the progress made thus far - will I be starting from scratch?
A System Rebuild will only rebuild the docker containers of any services you have installed. Your data will be safe.
Total ram of less than 2GB is not ideal - I know it claims you have a lot of ram free since you only have bitcoin installed, but there is some bug on low-memory Pis <8GB that we’ve been unable to explain which makes bitcoin absolutely crawl. You may just be experiencing this. Perhaps try our x86_64 build and some cheap repurposed hardware or upgrade your Pi to an 8GB. Unfortunately that’s the best advise we can give after spending lots of time to try to figure out what’s wrong to no avail. It’s one reason we’re moving away from the Pi for builds w/ external drives, and don’t support the <8GB pi anymore.
Thanks once again for a very helpful reply George. I’d like to consider a X86_64 self build but I don’t seem to find any instructions / video on the hardware requirements and build instructions. I’m about as far from being proficient at tech builds as can be (the Raspberry Pi build was a challenge) Would you mind guiding me to any relevant information/instructions I may be missing?
Ok. So I found some resources on x86_64 option - As I understand it, it’s not a “build” per se. Rather it involves flashing a copy of Start9 OS on a pre-built mini PC? And I noticed a forum covering different types of compatible hardware ranging from Lenovo Thinkcentre to HP Mini. So as I understand all that would be required would be to obtain one of these with the adequate memory and storage, flash a copy of Start9 on to it, and go from there??
Yes. This is correct.
Facing same problem here. I did a rebuild time ago: no improvement.
0% packet loss, ping around 25/45ms, download around 100Mbs
Last ping check results:
“271 packets transmitted, 271 received, 0% packet loss, time 270396ms
rtt min/avg/max/mdev = 25.504/32.620/116.403/8.612 ms”
Hardware: a laptop with Intel(R) Pentium(R) CPU N3540 @ 2.16GHz / 2TB ssd / 8gb ram.
Services installed: Bitcoin Core, Electrs, Jam, LN Terminal, LND, Mempool, Ride The LN.
There are moments where my node is just unusable, bitcoin core fails and everything falls apart (sometimes only the synced section, sometimes both rpc and synced, sometimes remains 1 or 2 blocks behind). This is a big problem, did someone of you guys are facing this issue? Did you find a way to fix it?
Unfortunately, your build is only slightly more powerful than a RasPi. Running an extensive Bitcoin stack is likely to only be viable for a short time with this hardware due to the growing resource consumption.
It’s also worth mentioning for any visitor to this thread that a healthcheck timeout is not necessarily indicative of a real problem. We have a bug that causes this message to be overly-aggressive, and it will be fixed in StartOS v040. You can check your Bitcoin logs to verify if you have a real problem or not.