Has anyone seem this recent discussion about Umbrel[1] particular line of code:
--snip--
--snip--
Indeed it has some privacy concern. Although this concern could be solved easily if Umbrel dev add feature to opt-out/disable automatic check.
Topic: Umbrel — Discussion, issues, solutions (Read 2167 times)
Indeed it has some privacy concern. Although this concern could be solved easily if Umbrel dev add feature to opt-out/disable automatic check.
umbrel presented the new umbrel home and umbrelOS 1.0 a few days ago.
the umbrel home is equipped with the new Intel n100 chip, a faster 16gb ram and a 2tb ssd and is also said to be 5x faster compared to the raspberry pi 4 and 2x faster compared to the raspberry pi 5 (see images below)
and the new umbrelOS 1.0 will be available from march 18, 2024 for umbrel home and rpi users. from april 2024 also for ubuntu and debian users.
https://umbrel.com/umbrel-home
https://umbrel.com/umbrelos
the umbrel home is equipped with the new Intel n100 chip, a faster 16gb ram and a 2tb ssd and is also said to be 5x faster compared to the raspberry pi 4 and 2x faster compared to the raspberry pi 5 (see images below)
and the new umbrelOS 1.0 will be available from march 18, 2024 for umbrel home and rpi users. from april 2024 also for ubuntu and debian users.
https://umbrel.com/umbrel-home
https://umbrel.com/umbrelos
So Umbrel finally follow their competitor (e.g. MyNode and Casa) by releasing their branded hardware. For ready to use hardware, it's not that expensive although it's weird it doesn't seem to have any video port (e.g. HDMI).
umbrel presented the new umbrel home and umbrelOS 1.0 a few days ago.
the umbrel home is equipped with the new Intel n100 chip, a faster 16gb ram and a 2tb ssd and is also said to be 5x faster compared to the raspberry pi 4 and 2x faster compared to the raspberry pi 5 (see images below)
and the new umbrelOS 1.0 will be available from march 18, 2024 for umbrel home and rpi users. from april 2024 also for ubuntu and debian users.
https://umbrel.com/umbrel-home
https://umbrel.com/umbrelos
the umbrel home is equipped with the new Intel n100 chip, a faster 16gb ram and a 2tb ssd and is also said to be 5x faster compared to the raspberry pi 4 and 2x faster compared to the raspberry pi 5 (see images below)
and the new umbrelOS 1.0 will be available from march 18, 2024 for umbrel home and rpi users. from april 2024 also for ubuntu and debian users.
https://umbrel.com/umbrel-home
https://umbrel.com/umbrelos
Indeed it has some privacy concern. Although this concern could be solved easily if Umbrel dev add feature to opt-out/disable automatic check.
That's true and one user already opened this issue[1] on Umbrel GitHub page. So far their CTO hasn't said that they will implement that feature in the future, but the debate is still fresh. I wouldn't be surprised if they do end up implementing that feature in the future in order to not compromise their stance on being a privacy focused solution...[1]https://github.com/getumbrel/umbrel/issues/1680
Has anyone seem this recent discussion about Umbrel[1] particular line of code:
Their CTO already came forward with a thread about the code[2] but some twitter users are still not happy about it. Umbrel CTO said that it is mostly used to keep track of what OS version the user is running but it still leaves a lot to be discussed about. Some users even asked why wasn't the ping being made by TOR (such as Zack[3] from Foundation Devices) but it seems that won't work[4] according to their CTO.
[1]https://nitter.lacontrevoie.fr/re_tweet/status/1682101313587363841
[2]https://nitter.lacontrevoie.fr/lukechilds/status/1682314300743118851
[3]https://nitter.lacontrevoie.fr/zachherbert/status/1682356123691630593
[4]https://nitter.lacontrevoie.fr/lukechilds/status/1682376830580588544
Their CTO already came forward with a thread about the code[2] but some twitter users are still not happy about it. Umbrel CTO said that it is mostly used to keep track of what OS version the user is running but it still leaves a lot to be discussed about. Some users even asked why wasn't the ping being made by TOR (such as Zack[3] from Foundation Devices) but it seems that won't work[4] according to their CTO.
[1]https://nitter.lacontrevoie.fr/re_tweet/status/1682101313587363841
[2]https://nitter.lacontrevoie.fr/lukechilds/status/1682314300743118851
[3]https://nitter.lacontrevoie.fr/zachherbert/status/1682356123691630593
[4]https://nitter.lacontrevoie.fr/lukechilds/status/1682376830580588544
My followup: IBD finished after little less than 95h on a Raspi 4B with 8GB RAM and a 1TB SATA SSD (Crucial MX500). Sync of Electrs took less than a day (sorry, don't know yet how to access the relevant log file for electrs in the container; I haven't seen anything else useful mapped to the "outside" filesystem. Before I installed the Electrs app, I reduced dbcache=512 for Bitcoin Core.
Roughly 4 days for the IBD with a Raspi 4B 8GB RAM via Tor isn't bad, so far, actually I'm quite pleased.
Roughly 4 days for the IBD with a Raspi 4B 8GB RAM via Tor isn't bad, so far, actually I'm quite pleased.
My Umbrel test node (it's on Mainnet, I use it only as a node distribution test and to host a few of the Umbrel apps, no serious node business) crashed with Umbrel 0.5.3 running. It seems that there's no reliable docker stuff garbage collector. The system SSD ran full as I don't watch it daily. I observed this before that space did run low slowly, but of course faster as the blockchain grows. An Umbrel reboot always reclaimed the free space on the SSD. Frankly, I don't understand why this happens at all.
Anyway I deleted following apps from my Umbrel: Bitcoin Core, LND, Electrs, Bitcoin Explorer, Mempool.space and left Snowflage and Uptime Kuma running. Updated Umbrel to 0.5.4 and re-installed only the Bitcoin Core app for academic reasons. I want to test how fast Umbrel does an IBD with a tweaked setting to hopefully boost IBD speed.
My Raspi 4B has 8GB RAM, a 1TB SSD and is connected via LAN to my router which handles a reliably full 100MBit down stream internet connection. I set DB cache for Bitcoin Core to use 5120MB (this is my tweak for IBD). Though, Umbrel passed only the parameter dbcache="4883" to Core and doesn't show this in the Advanced Settings of Core. I don't like such kind of paternalism. Better show if you overdo something.
Now after roughly 50h IBD the Umbrel node is at ~64%. I'll report back when IBD has finished. When Bitcoin Core has fully synced I will set DB cache back to standard and throw in Electrs, Mempool Explorer and Mempool.space (no LND this time as I'm not going to use LN on Umbrel).
Maybe I will also repeat the IBD with Bitcoin Core and Electrs running concurrently, again with a high DB cache value. I want to see if and how much difference there is running it in sequence or in parallel. (I was thinking that Electrs might eat up too much of a chunk of RAM so that there's less available to max. dbcache value. To my understanding an as high as possible dbcache value helps to speed up IBD.)
My internet connection is definitely no bottleneck here as it is most of the time only utilized about 20-30% of max. bandwidth during IBD and when blocks flow in.
Anyway I deleted following apps from my Umbrel: Bitcoin Core, LND, Electrs, Bitcoin Explorer, Mempool.space and left Snowflage and Uptime Kuma running. Updated Umbrel to 0.5.4 and re-installed only the Bitcoin Core app for academic reasons. I want to test how fast Umbrel does an IBD with a tweaked setting to hopefully boost IBD speed.
My Raspi 4B has 8GB RAM, a 1TB SSD and is connected via LAN to my router which handles a reliably full 100MBit down stream internet connection. I set DB cache for Bitcoin Core to use 5120MB (this is my tweak for IBD). Though, Umbrel passed only the parameter dbcache="4883" to Core and doesn't show this in the Advanced Settings of Core. I don't like such kind of paternalism. Better show if you overdo something.
Now after roughly 50h IBD the Umbrel node is at ~64%. I'll report back when IBD has finished. When Bitcoin Core has fully synced I will set DB cache back to standard and throw in Electrs, Mempool Explorer and Mempool.space (no LND this time as I'm not going to use LN on Umbrel).
Maybe I will also repeat the IBD with Bitcoin Core and Electrs running concurrently, again with a high DB cache value. I want to see if and how much difference there is running it in sequence or in parallel. (I was thinking that Electrs might eat up too much of a chunk of RAM so that there's less available to max. dbcache value. To my understanding an as high as possible dbcache value helps to speed up IBD.)
My internet connection is definitely no bottleneck here as it is most of the time only utilized about 20-30% of max. bandwidth during IBD and when blocks flow in.
yesterday umbrel introduced/released its new version. with verion 0.5.4 there is a new migration assistant to seamlessly transfer all apps and data from a rpi based umbrel to umbrel home. in addition to this improvement, there are several performance enhancements, bug fixes and subtle improvements, such as the option to update your display name via the settings.
to update your umbrel node to the new version, just log in and click on the update button - thats all
https://twitter.com/umbrel/status/1671513852570206225
to update your umbrel node to the new version, just log in and click on the update button - thats all
https://twitter.com/umbrel/status/1671513852570206225
which mobile lightning wallet app do you use to connect to your own umbrel ln-node?
i used zeus back then but somehow the new version of zeus doesn't recognize the qr-codes of the ln-node anymore
i used zeus back then but somehow the new version of zeus doesn't recognize the qr-codes of the ln-node anymore
since i made a backup of all the payment channels back then, i deinstalled the ln-node and then reinstalled it.
now i can access them again and all is well
now i can access them again and all is well
From what I can see and remember it looks like a corrupted database.
Did it crash and reboot that you know of?
And do you have a backup?
Reach out to the Umbrel telegram group there might be a way to rebuild but I usually just wipe and restart.
Don't know if you have any funds in there so that may change the way you deal with this.
-Dave
Quote
lnd_1 | panic: freepages: failed to get all reachable pages (page 50741: multiple references)
Did it crash and reboot that you know of?
And do you have a backup?
Reach out to the Umbrel telegram group there might be a way to rebuild but I usually just wipe and restart.
Don't know if you have any funds in there so that may change the way you deal with this.
-Dave
turned off my umbrel node for a few minutes and then started it up again. unfortunately i still can't connect into my ln-node
various browsers were also tried - without success
here the log after i started the troubleshooting - maybe someone of you can find a error...
various browsers were also tried - without success
here the log after i started the troubleshooting - maybe someone of you can find a error...
Code:
lightning
Attaching to lightning_lnd_1, lightning_tor_server_1, lightning_app_proxy_1, lightning_app_1, lightning_tor_1
lnd_1 | 2023-02-25 17:03:02.098 [INF] RPCS: gRPC proxy started at 0.0.0.0:8080
lnd_1 | 2023-02-25 17:03:02.098 [INF] LTND: Opening the main database, this might take a few minutes...
lnd_1 | 2023-02-25 17:03:02.098 [INF] LTND: Opening bbolt database, sync_freelist=false, auto_compact=false
lnd_1 | panic: freepages: failed to get all reachable pages (page 50741: multiple references)
lnd_1 |
lnd_1 | goroutine 196 [running]:
lnd_1 | go.etcd.io/bbolt.(*DB).freepages.func2()
lnd_1 | go.etcd.io/[email protected]/db.go:1056 +0x90
lnd_1 | created by go.etcd.io/bbolt.(*DB).freepages
lnd_1 | go.etcd.io/[email protected]/db.go:1054 +0x144
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
tor_1 | Feb 25 16:43:39.000 [notice] Bootstrapped 55% (loading_descriptors): Loading relay descriptors
tor_1 | Feb 25 16:43:41.000 [notice] Bootstrapped 62% (loading_descriptors): Loading relay descriptors
tor_1 | Feb 25 16:43:42.000 [notice] Bootstrapped 67% (loading_descriptors): Loading relay descriptors
tor_1 | Feb 25 16:43:43.000 [notice] Bootstrapped 75% (enough_dirinfo): Loaded enough directory info to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 80% (ap_conn): Connecting to a relay to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 85% (ap_conn_done): Connected to a relay to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 89% (ap_handshake): Finishing handshake with a relay to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 90% (ap_handshake_done): Handshake finished with a relay to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 95% (circuit_create): Establishing a Tor circuit
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 100% (done): Done
Attaching to lightning_lnd_1, lightning_tor_server_1, lightning_app_proxy_1, lightning_app_1, lightning_tor_1
lnd_1 | 2023-02-25 17:03:02.098 [INF] RPCS: gRPC proxy started at 0.0.0.0:8080
lnd_1 | 2023-02-25 17:03:02.098 [INF] LTND: Opening the main database, this might take a few minutes...
lnd_1 | 2023-02-25 17:03:02.098 [INF] LTND: Opening bbolt database, sync_freelist=false, auto_compact=false
lnd_1 | panic: freepages: failed to get all reachable pages (page 50741: multiple references)
lnd_1 |
lnd_1 | goroutine 196 [running]:
lnd_1 | go.etcd.io/bbolt.(*DB).freepages.func2()
lnd_1 | go.etcd.io/[email protected]/db.go:1056 +0x90
lnd_1 | created by go.etcd.io/bbolt.(*DB).freepages
lnd_1 | go.etcd.io/[email protected]/db.go:1054 +0x144
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_1 | Waiting for LND...
app_1 | Checking LND status...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
app_proxy_1 | Validating token: dab2fdf87bd9 ...
tor_1 | Feb 25 16:43:39.000 [notice] Bootstrapped 55% (loading_descriptors): Loading relay descriptors
tor_1 | Feb 25 16:43:41.000 [notice] Bootstrapped 62% (loading_descriptors): Loading relay descriptors
tor_1 | Feb 25 16:43:42.000 [notice] Bootstrapped 67% (loading_descriptors): Loading relay descriptors
tor_1 | Feb 25 16:43:43.000 [notice] Bootstrapped 75% (enough_dirinfo): Loaded enough directory info to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 80% (ap_conn): Connecting to a relay to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 85% (ap_conn_done): Connected to a relay to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 89% (ap_handshake): Finishing handshake with a relay to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 90% (ap_handshake_done): Handshake finished with a relay to build circuits
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 95% (circuit_create): Establishing a Tor circuit
tor_1 | Feb 25 16:43:44.000 [notice] Bootstrapped 100% (done): Done
somehow i can't connect to my lightning node since a few days. when i click on the ln icon in the umbrel webif a new tab window opens as usual but the ln node can't be opened or loaded.
the website tries to build up but nothing happens...
the website tries to build up but nothing happens...
Basic diagnostics that I know you know but....
Did you reboot the Umbrel node?
Try a different browser?
Other then that how is the node behaving?
Remember, Umbrel puts everything into docker containers more or less. One of them could just be having an issue.
-Dave
somehow i can't connect to my lightning node since a few days. when i click on the ln icon in the umbrel webif a new tab window opens as usual but the ln node can't be opened or loaded.
the website tries to build up but nothing happens...
the website tries to build up but nothing happens...
since yesterday you can run a nostr relay on umbrel.
with the damus app (which you can dl in the umbrel app store) or another nostr client your complete nostr acitivity is backed up and in real-time
https://twitter.com/umbrel/status/1622591145481236480
with the damus app (which you can dl in the umbrel app store) or another nostr client your complete nostr acitivity is backed up and in real-time
https://twitter.com/umbrel/status/1622591145481236480
the tor network can now be deactivated in the umbrel.node as of now
after logging in to your node, just click on the 3 dots '...' in the upper right corner and then choose 'advanced settings
after logging in to your node, just click on the 3 dots '...' in the upper right corner and then choose 'advanced settings
just before the turn of the year and just in time for christmas, there is now with firefox, wikijs, n8n, audiobookshelf, remmina and snapdrop new 6 self hosted apps at the umbrel store
It's absolutely not true that it takes so long to synchronize the node
There are many factors that can slow down the process. One's the CPU as you correctly pointed out. But my CPU wasn't that bad, it was a Raspberry Pi 4. Once I installed Raspibolt syncing went faster. As I've already said in this thread, the reason why it was taking weeks, was that it didn't establish connections. I barely had 1-2 connections, and sometimes even zero for a few minutes. Downloading using Tor didn't slow down the process so much as you think it did.
This is pretty much the reason I uninstalled Umbrel. It didn't allow me sync fast. It takes about twice to sync that way. Also, I had very poor connectivity. There were times that it had 0 outgoing connections, and paused syncing.
It's absolutely not true that it takes so long to synchronize the node, actually with my rasp it took just 6 days which is average, but it doesn't depend on your connection or the tor network. During block validation it is the CPU that acts as a bottleneck so if you don't have great hardware it could take a long time or if you have great hardware it will take much less.
is there actually already a possibility for umbrel, where tor can be deactivated and one acts as a full-fledged network node?
How is it not a full-fledged network node if it is configured to connect to the Bitcoin network over Tor?In case this is an XY Problem: what are you trying to accomplish?
Jump to: