Bitcoin Forum

The Raspberry Pi 4 is here, if someone gets one, could you post performance input including:

1) Time for the Bitcoin wallet to open
2) Time to download and sync the entire blockchain
3) Time to sync exiting copy of the blockchain

+ any other relevant performance info.

State your OS and your storage device.

Interested in the 4 GB version..

Dont got one yet but the raspberry pi can handle 1gib connections and with the 4gib of memory it will boil down to your hard disk speeds.

CPU was the biggest bottleneck when it came to synchronising with the network. Depending on the task, the CPU seems to be up to 3 times faster which is a huge improvement. No one has tested it yet with Bitcoin Core.

It says below im looking for 4 GB. I don't get why you would get any less... if you wanted to use it as a desktop computer as well you want at least 4. If you wanted to run a Windows node for some reason definitely impossible without less than 4 GB or at least not without a big PITA.


Not only the CPU but the Ethernet was capped at 300mpbs which was a big bottleneck if you had high speed optical fiber of 1 GBPS which you can take advantage of now with the 4 version. 4 GB doesn't hurt. Ideally we want 8 for the next release.

I can't say this strongly enough.

IF you really want to run a node on an RPi think about the following.

If at all possible do not run it on the SD card. Get an external drive and make it an SSD

If you can't or don't want to. GET A GOOD SD CARD. There are good cheap ones out there but they are few and far between, spend the money and get a real name brand.
The cheap ones degrade quickly and core does a lot of reads & writes.

-Dave

Despite of decreasing SSDs prices, they are still quite expensive if you compare them to HDDs (especially used ones). If he is willing to use a Raspberry Pi then he probably wants to spend as little money as possible. Also, high capacity SD cards are way more expensive than SSDs. Thank you Dave for warning us against something that nobody would do.

SD Cards are trash. If you wanted to stay digital then SSD is the way to go... but most likely you will be forced to run pruned mode pretty soon since the big TB drives are pretty expensive.

I don't see why not just you couldn't stay with HDD. Sure it's slower, but if you want to run a full node (no pruned mode) then it makes sense to get a 4TB one and forget about hitting a limit for years.

I have started ditching HDDs mostly because they are much slower than SSDs (especially NVMe ones) and they are not shockproof. I wouldn't use an SSD for a budget build like the one with a Raspberry Pi, but spending $120 on a 1 TB SSD is a good deal if you are setting up a server which is supposed to handle a few different tasks beside running Bitcoin Core and LND.

Samorai retweeted a tweet of a user (https://twitter.com/erik_jacobi/status/1152511845481291776) who is currently synchronising a full Bitcoin node on the latest Raspberry Pi. I will keep you updated on the sync time but the problem is that this users is using an SSD which is going to affect the performance. I have switched to NVMe drive from HDD and the difference was huge.

I run a full node on Rapi3+.
I followed Stadicus Guide and translated it in Italian (in signature) to help other running their own node.

For a series of mishaps my block.dat got corrupted, so I had the blockchain download in my SSD and I had the Raspi to validate it.
It took more than a month to fully validate the blockchain back in february.

Of course this is not what it is meant to be, but of course improved CPU performance and RAM are a more than welcome upgrade in this case.
So I second OP request to know more on validating/syncing blockcahin.

Re:SSD, in my experience they are not meant to be used as a constant R/W devices like an hard disk, so I have been told their use as mass storage is not optimal and quite prone to errors and data corruption.

Waiting a little bit to put my hands on a Raspi4 because of I heard of a few minor design flaws in the board.
https://arstechnica.com/gadgets/2019/07/raspberry-pi-4-uses-incorrect-usb-c-design-wont-work-with-some-chargers/
Nothing too relevant,. but waiting they get rid of their stocks and design a new board version.

The author of the tweet I mentioned earlier still hasn't posted the results yet.


That's how I would describe SD cards. SSDs improved a lot in the past few years. They have become affordable and reliable. Some SSDs last longer than advertised, but every device is prone to failure at some point. Why would anyone use them if they were not meant to be used for constant reads and writes?

what to look for:

• NAND cell tech - SLC is best, MLC 2nd, TLC 3rd and QLC 4th in reliability/longevity. Pricing corresponds; SLC based SSD devices are expensive, QLC cheap
• Over-provisioning - a larger over-provided pool of flash cells will improve reliability and reduce performance. Some SSDs let you reprogram the amount of over-provisioning via the firmware

there was another factor also, but I've forgotten it for now

SLC SSDs are mostly used for server applications and they are going to be harder to obtain for an average consumer. MLC drives are still widely available. QLC drives seem to be priced about the same as TLC ones (at least in my country) while being less durable.



Wasn't it Total Bytes Written (TBW) by any chance? If TBW is exceeded then the warranty is void, but the drive still can work for a decent amount of time. Also, the controller is a very important factor. If it breaks then all of the data becomes unrecoverable in most cases.

Totally agree.
System must be engineered to avoid bottleneck.
This  is what I tried to avoid setting up my Bitcoin node guide (in signature)as you can see if you check hardware page (https://github.com/Fillippone/NodoBitcoinforDummies/blob/master/02.Preparativi.md)

So, the RPi 4 generates a crap more heat from it's CPU then previous models.
Some people are having thermal throttling in under 5 minutes of heavy use.
So I give you the 52Pi Ice Tower

https://www.seeedstudio.com/ICE-Tower-CPU-Cooling-Fan-for-Raspberry-pi-Support-Pi-4-p-4097.html (https://www.seeedstudio.com/ICE-Tower-CPU-Cooling-Fan-for-Raspberry-pi-Support-Pi-4-p-4097.html)

Toms Hardware Review Link:
https://www.tomshardware.com/reviews/52pi-ice-tower-raspberry-pi-4-cooler,6259.html (https://www.tomshardware.com/reviews/52pi-ice-tower-raspberry-pi-4-cooler,6259.html)

If you want something a bit more sane:
https://shop.pimoroni.com/products/fan-shim (https://shop.pimoroni.com/products/fan-shim)

I'm not just posting this for laughs, I had my 4 thermal throttle while nothing but bitcoind was running.

-Dave

Well this is something undoubtedly serious and needs to be taken into account while tinkering with a Raspi4 to run a Bitcoin full node. I like them quiet and reliable, tossing on desks and basically expecting them to work indefinitely. Of course heat is a very dangerous element to have on such a tiny device.
Thanks for pointing us in the right direction.

Yeah, I was trying to figure out why my node stopped responding now and then. And someone else on another forum was having issues with theirs overheating.
So I put a small fan on it and went to look around, and saw that it is happening to a lot of people. Someone someplace cut some corners that they should not have when designing the 4

-Dave

Rasp pi are simply not made for this task.

Cheap used Lenovo tiny PCs do a better job.

Dell has a good small unit. Buy a used one. Many will hold a nvme and a ssd.

I use Mac mini’s 2012 model and or 2014 models.

I clone a backup os with a wallet . And away I go.

sort of disagree (sort of)

storage is really what holds the Pi back these days. CPU did before, but not since the rPi 2. You can get around the SD card issues by using the most modern rPi's (3B+), as they allow you to boot from a SATA disk, bypassing SD cards altogether.

But really, USB storage, SATA or not, is less than ideal. You can get a Rock64 pro single board for less than $100. You can also get an NVMe or SATA PCI-E expansion card for the Rock64 pro (it has only 1 slot). And there's a 4GB version, you could run a lightweight hypervisor on that, and get a securely compartmentalized home Bitcoin node, + run other stuff in different VMs (it'd have to be without a desktop GUI, naturally)

I have been running my personal BTC+LN node on a Raspi3+ for months now (link to my guide to do so in signature) and I hadn’t a single issue.
I agree USB connections are “ shaky” in theory compared to SATA, but, touch wood, I hadn’t any issue so far.
Of course my node is “for fun”, not holding millions there. If I had to store a lot of money I would walk the extra steps.

a defining characteristic of unreliability is: it's unreliably unreliable :D

a USB disk crashing your Pi node is ok when you're only an hour away from where you're running the node. But if you're on vacation hundreds or thousands of miles away, you may find that annoying.

I misread your previous comment “hundreds of thousands miles away” and I was more worried about your holiday habits than USB reliability ! ;)
I get the point: reliability can be issue also with small amounts.

I both love and hate RPi units.
To say they are not up to the task is not quite fair, they are up to it but barely.
The 4 (if it was not for the heat issue) is more then up to it.
It's when you are using your RPi as a node & a pi-hole & open VPN front end for your home network all at once you really see problems.
I am actually using the one I got from from the Avalon deal that never happened as a smart controller in the condo.

I have an PowerEdge server running Hyper-V that handles most of my "heavy lifting" but I also have many a RPi sitting around doing stuff too.

The nice thing about them is to play they are cheap and simple.
And, you can power it down from bring your node, swap in a different SD card and now it's back to being your VPN server, power it down again and put in another SD card and it's doing something else.

Do I 100% rely on it. No.

As with everything, YMMV.

-Dave

Yeah  I use them..  

 but used mac minis are pretty much bullet proof in comparison.
you can do linux if you chose.

and

small dell
small lenovo can be found cheaply.


here  are some low cost lenovo


https://www.ebay.com/itm/Lenovo-ThinkCentre-M92p-Tiny-i5-2-9GHz-8GB-500GB-Win-10-Pro/332379682378?

If you really look  sometimes you get them at 100 bucks

some cheap mac minis

https://www.ebay.com/itm/Apple-Mac-Mini-Desktop-2012-i5-2-5GHz-4GB-500GB-HDD-C-Warranty/273868158121?

when lucky I get them at 100-125

use this ssd dock

https://www.ebay.com/itm/Blackmagic-Design-MultiDock-2-4-SSD-Port-Thunderbolt-2-Loop-Through/264393927328?

I got mine for 265 on a close out.

they all just work so much better then the rasp pi's  and they don't use a lot of power.

the toughest one to get a bargin  on is the dock from blackmagic.  But it is a 10 year to 15 year item.

The problem is validating the blockchain is a pretty single core intensive task so Raspberry type of devices are always going to be very limited in this area unless they can somehow keep pushing clock speeds without sacrificing other things.

Perhaps the best candidate is the Lenovo x200, at 2.4ghz it should be enough, then you've got the fact that you can Libreboot it which is pretty cool considering you are removing all the proprietary blobs that come with the clusterfucks formerly known as BIOSes.

Fyi. in case others are still interessted.

I got a fresh raspberry pi 4 Model B (4GB), connected a SSD using USB-SS and started sync'ing on a fast internet connecting.

After just a few hours I'm at 16% progress with >40GB downloaded and estimated 24 hours left.

Compared to older raspberries, the 4 are blazing fast  ;D

Yeah, CPU temp and eventful heat induced throttling are something I am very interested into.
Blockchain verifying is a CPU RAM intensive tasks, so heat should be a thing to care about.

I left it sync'ing at my workplace. Will be interesting checking the status tomorrow to see if it could keep up the pace with the larger blocks.
I have a small passive heatsink on the cpu. It gets hot but sync'ing seems to be mostly a single core activity so it doesn't feel critically hot. Don't think it throttled down.
RAM usage was in the range of 1.3 GB when I checked, so don't go for the 1GB variant. It's running raspbian buster lite.
There's plenty of bandwidth to the external SSD through the super-speed USB connection. Also the gigabit LAN is no bottleneck.

Oh such a nice thread, my RP 4B 4gb is on its way to me so when it comes (and hopefully I'll have enough time to play with it) I'll post some stats as well. Waiting to get FTTH sometime by the end of the month, then I can give real values.

I have some old laptops laying around - i3,i5, maybe I can make a comparison between them.

I have the feeling that heat sink are a unwelcome news for Pi4 owner, and the result of a poor design choice from Raspberry.
Heat sinks are bad for at least three reasons:

• They are dust collectors and need to be cleaned to maintain efficiency, while this HW should be thrown everywhere and keep functioning for years to come without being touched.
• They prevent the use of external cases to keep the board cleaned (see above)
• They are aesthetically horrible (If you don’t agree with the “throwaway statement” and you think you want to see your node)
• I have a bad feeling for a heat sink being necessary on an hardware that it is used for his reliability, when delivering an unrequested (yet welcome) performance boost.

I believe our ISP or other internet infrastructure has limited the throughput to my raspberry. Maybe someone thought +100GB/day is excessive for a single device  ;D
The sync speed dropped significantly over night. So did CPU usage (and temperature). As far as I can tell, it's limited by the network as neither CPU or disk are loaded.

Screenshot:
http://too-hot.dk/storage/temp/rpi_btc.png

But 45% done after about 20 hours are not bad either :-)

Heat management has been a major concern with RPI4. Unfortunately it's limited by technology. You only get a certain amount of MIPS per watt at a certain technology node. (btw. I used to work for Intel).
RPI4 manages it pretty well i.e. throttles down the clock when temperature exceeds 80C - it's built for operating at those temperatures (I hope). So you can either accept a slight decrease in performance at high workloads or add some heat dissipation capabilities. Or play around with under-volting/under-clocking if that's your thing.
I just got a small heatsink for now (http://too-hot.dk/storage/temp/rpi_heatsink.png) which helps quite a bit. As you can see, it's at 60C with the moderate load. It still reaches 80C at full load though even with this heatsink (however, that requires more active processes than 'just' bitcoin).
There are a lot of cool cases for the RPI4 comming out, where the case is made of aluminum coupled to the CPU.

We indeed need cool cases.

Sorry, couldn't resist a silly comment.

Will try to order one or both of these 'cool'  :P cases from china. There are also options with fans but I really prefer passive cooling for something that is intended for 24/7 use.
https://www.aliexpress.com/item/4000056606252.html
https://www.aliexpress.com/item/4000095452880.html

other performance bottlenecks are:

• the dbcache setting. With 4GB RAM, setting this to 1GB (i.e. dbcache=1024) should be fine
• Despite all the RAM, setup a new swapfile (2-4GB) on the SSD, then after turning that new swapfile on, turn off the old swapfile on the slow SD card
• Use bitcoin 0.18.0+ to get better block processing performance (0.18.0 had a double digit performance increase in the database handling code)

Yes, so long as you don't use a crap drive or one that is using a crap USB -> SATA chip.

I work on too much low end hardware, and some of the cost cutting things are really irksome.

Some of the newer low end external SSDs have worse performance then an older spinning drive.


Make sure that there is actual contact between the case and CPU in the proper area. Cheap cases suck.
Just scrapped a bunch of aluminum cases for the RPi3 that had enough space between the heatsinnk that was part of the case and the CPU itself that the thermal pad I put on the CPU was untouched on top :(

-Dave

There's also a great overview on cases/cooling here: https://www.martinrowan.co.uk/2019/09/raspberry-pi-4-cases-temperature-and-cpu-throttling-under-load/

Passive cooling i.e. natural convection does indeed require a really huge heatsink compared to forced air-flow cooling. To be effective you basically need the entire case to be a heatsink.

If you have the data it is fairly easy to calculate the temperature rise of a heatsink given the required power dissipation. Or at least getting a rough estimate.
It might be difficult finding data for a Chinese raspberry case, but let's say we want a heatsink to be 30C above ambient then you would need one with a thermal resistance of 3 degree C per watt at natural convection.
A random example of a heatsink specified for 3 C/W @ natural (not suitable for RPI but just to provide a hint about size) : https://www.digikey.com/product-detail/en/wakefield-vette/401K/345-1048-ND/340344
On top of this comes the thermal resistance in the interface between the chip and the heatsink, which adds to the temperature rise of the silicon. On the other hand, a lot of the power dissipation is caused by other chips on the RPI board than the CPU.
Sorry for blabbering. Just happened to be an electrical engineer doing stuff like that for a living  ;D

Btw. I've plugged the RPI to my own LAN. It's sync'ing much faster again. Should be done sometime tomorrow :-)

Did the sync ever finish OK?
Also wondering if you ordered the cases or are doing anything else with the cooling. I'm still kicking around ideas for mine, none look great.

-Dave

Yes it finished. Not without problems though  ::)

I naively thought the blockchain would still just fit on my 250GB SDD but turned out I was about 20 gig short.
Then I moved the data to a larger 5400 rpm HDD which of course is awfully slow at sync'ing.

Then then project turned into a filesystem experiment. I re-formatted the 250GB SSD to a btrfs filesystem with zlib compression enabled and copied the blockchain back.
Even though the blockchain is poorly compressible it still compresses to about 80% i.e. freeing up sufficient space with some 10GB to spare after sync'ing was done.
Naturally, the compressed SSD is also significantly slower than uncompressed it is still much faster than the HDD. I guess I just need to order a new disk..

Now I'm fiddling around with raspiblitz to get some hands-on with lightning  ;D

I ordered one of those china aluminium boxes - will see how it fits and whether the thermal interface can be done properly. But as mentioned, it is not really a huge problem. It will throttle down if it gets too hot and normally it won't even get that hot when just staying in sync.

Very interested in temperature and eventual thermal throttling.
Heard of many instances of thermal throttling of R4 just being sitting idle (probably due to very poor air circulation/high ambient temp).

Very interesting thread here:
https://www.raspberrypi.org/forums/viewtopic.php?f=28&t=243500&sid=6c9afc7c1a072ffb808168ebaa95a7ea&start=525#p1536192

https://www.raspberrypi.org/forums/download/file.php?id=32041

Wondering if just running bitcoind + LN hit the 80 C° threshold.

Can you mention how much time did it took to finish the sync? Does Bitcoin Core keep such a statistic? this would be interesting to have so you could keep track of it even if you close the device you are using. Just count the time it takes until you reach the first ever sync, this would be extremely useful for the database im doing. Right now im running a bunch of different devices and keeping track of it all with chronometers because as far as I know the client doesn't have anything built in to do this.

I wonder if there is a massive difference Linux vs Windows but that would be x2 the time it will take for this to finish... I think it's best to put this effort in Linux for now.

I can't say for sure as I had some troubles with the last part of the sync and did some experiments etc. But at least, the first half of the blockchain (i.e. +100GB) was downloaded within the first day.
With optimum conditions (fast network and SSD drive) I believe a  clean sync should easily complete within two days on the RPI4B.

I now also got raspiblitz (Bitcoin + LND) up and running :-)
With the small passive heatsink temperature settles at about 68C with some open connections...


               RaspiBlitz v1.3  Blitz99
               bitcoin Fullnode + Lightning Network
        ,/     -------------------------------------------
      ,'/      CPU load 3.85, 3.75, 2.71, temp 68°C 154°F
    ,' /       Free Mem 3392M / 3906M  HDDuse 262G (95%)
  ,'  /_____,  ssh admin@192.168.0.164 d8.3MiB u25.8MiB
 .'____    ,'
      /  ,'    bitcoin v0.18.1 mainnet Sync OK 100.00%
     / ,'      Public xx.xxx.xxx.xx:8333 17 connections
    /,'
   /'          LND 0.7.1-beta wallet 0 sat (+400000)
               0/0 Channels 0 sat 4 peers

I don't think you can extrapolate the remaining half of the blockchain by what amount of time it took to get the first half downloaded. Consider this:

https://jochen-hoenicke.de/queue/#0,all

Unfortunately there isn't older charts that collect this data that I know of, but by looking at that you can see the all time peaks of 2017 during the $20k bubble, that mix of legit traffic + Ver and Bitmain spamming makes that section of the blockchain specially slow to synch in my experience. Maybe it was because it's an HDD and you notice this section less with SSD. When I find some free time I will try for myself and monitor the full sync time and report back.

Received one of the cheap cases, the one with ribs. And as DaveF suspected, there were no contact between cpu and heatsink. There were about 1mm air gap 😕.
I cut a small piece of 1mm aluminum and fitted it with heatsink compound, which turned it into an ON OK solution 🙂.
Raspiblitz running at about 50C now.

Edit: Fixed typo.

https://66.media.tumblr.com/373cf42f3c3a47a6fcceb2d24b681cf5/tumblr_inline_nawycwntG91s1ve0y.jpg

What is the ambient around the case? Any real airflow? That's a 18C temp drop which is a fairly large percentage.

-Dave

There's 22C ambient. No airflow but it's in 'the open' (i.e. unrestricted convection) in the corner of an office:
http://too-hot.dk/storage/temp/raspi_case.jpg
Showing 52C after a couple of hours - haven't seen it higher than that yet. Dropped to 51C after taking the image.

You can even tell I'm currently using a rotary disc on the thermal  image ;D
http://too-hot.dk/storage/temp/thermal.jpg

I was expecting about 50C given a sizeable heatsink with good thermal interface. A bit of forced convection will really help, but I want a completely quiet solution and 50C is totally fine :-)

Oh, by the way. I removed the anodized surface of the area of the heatsink that evenually contact the cpu. Anodized aluminum is a really good insulator of both electricity and heat. Probably doesn't matter much at this relatively low thermal densitity though. But we recently had an issue at work with a heatsinks delivery that was inadvertently anodized at the contact area where we transfer in excess of 100W/cm2 from a power tranasistor. This caused an increase in die temperature of more than 10C so we had to rework those heatsinks by milling the surface.

Woo-hoo look what I just got in my mail today:
https://i.imgur.com/I0myvYS_d.jpg?maxwidth=640&shape=thumb&fidelity=medium

I want to play a bit with it first and later I'll move my full node to the Pi.
Have other projects on the way so I'll need to wait with the testing and benchmarking, but when that time comes I'll share my experience with you guys :)

Thanks to the Bitcoin I can play with my new toy, thanks Satoshi ;)

Have fun playing :-)
I need to get a second one now that I allocated mine to run a bitcoin node  ::)

Tried leaving it running with 100% load on all cores (cpuburn-a53). It stabilizes at 71C after about an hour. Damn hot to the touch but no clock throttling.
If 100% load was a 24/7/365 thing, then I would definitely go with a fan but being an intermediate condition it is fine.

bitcoind and lnd doesn't even seem to mind the cpu being kept busy :-)

I think you didn’t mentioned exactly the mode of the case.
Would you please tell us which “cheap China case” did you buy?
Would you recommend it, given you had to do so much DIY to make it operational?
Thanks!

It's this one https://www.aliexpress.com/item/4000056606252.html
If you are comfortable adding a small piece of aluminum/copper in between the gap, then it's great. But if you like something that just works then stay away from it.
It's really odd those Chinese manufacturers don't check these kind of things before manufacturing by the thousands...

Has anyone been having any issues with performance with the September 26 image?
I wiped both of mine for different reasons and downloaded the newest one and although I can't give any hard and fast numbers it just feels like there was a performance hit.
I am wondering if they did something to help with the heat issue.

No help on the RPi forums so I figured I would ask here.

Thanks,
Dave

I have one running as a pi-hole and open VPN server.
There is a definite slowdown with it, it's not a "real" number for the pi-hole in the fact that it went from .26 seconds to .4 for a lookup it's still a 50% increase.
* The .26 number was from a test a while ago and the .4 is from today so I can't say when that slowdown happened.

The open VPN has dropped max throughput ~10% from Tuesday, that is where I really noticed it.

I'm going to wipe and start again, but I was wondering if it was just me.

-Dave

I was near the end of downloading the blockchain data or blk?????.dat when it started getting "warning 52 of last 100 blocks have unexpected version". I don't know what to do at this point!  I was able to figure out that when my 256GB SanDisk card got full, it was able to restart when I -prune=550 !  When I did the prune the 1st time, it started at 2019 for redo the blockchain! Maybe when I do a GUI, I'll do it next time when the blockchain gets full again on 2nd Raspberry Pi 4!  I keep looking for commands that would eliminate "warning 52 of last 100 blocks have unexpected version" so finish the blockchain download or go to GUI to finish downloading the Bitcoin blockchain.  I might get a 512GB SanDisk next time that'll cover the blockchain data.  200GB is not really enough for Bitcoin blockchain data anymore!

That warning is benign. There is no way to get rid of the warning, and that the warning exists does not cause any problems. You can safely ignore it.

I would look for any changes made to the

• linux kernel image
• linux firmware
• startup config (so, /boot/config.txt and /boot/cmdline.txt)

...relating to this latest update to Raspian, that's where any potential problems may have crept in. roll back to the old version to see if you can confirm the issue, then report that online (raspberry forums are best IMO for that)

these Pi-style computing boards are really not designed to be always-on servers, so getting them to work that way takes some additional effort. It can be done though.

there's another possibility: wipe an SD card too many times (or write too much to it in general) and you'll begin to increase unreliable behavior from the card, it's low quality flash memory that deteriorates quickly. It's better to use a SATA disk over a USB hub, particularly with the Pi4, as the Pi4 has a new SATA-over-USB protocol to improve that kind of setup (your USB hub needs to support the new protocol though)


I'll add that this happens because some Bitcoin miners decided to mis-use the protocol in a way that boosts the performance of their mining rigs.

and also to reiterate: this does not cause any problem, it doesn't matter in reality

benign warnings have been removed from normal logging before (1 just recently), there's a case to do that for this warning, or at least to change the text from "warning" to something more appropriate



this, however, will cause you a problem. I think we reached the point where 256GB disks were unusable a while ago

I did ask in the forums, but it rapidly turned into a "must be you, can't be us" thing. Which I do accept as yes it might be my hardware, my config or something else but was looking to see if anyone else was having the issue. There are scattered reports of people seeing the same issue but not a lot so it's not just me. Made me wonder if other people were having the same issue and just did not notice. Only reason I did was that I was moving a large file and I saw the speed being less then what it was. Has it not taken 1 hour on Monday and 1 hour 12 minutes on Friday I would have never seen it.



Not sure where you got that idea from. Yeah, if you are putting them in a always running at max CPU / RAM you need additional cooling on the chips and possibly some other small changes. But for small embedded machines they are perfect. I have the following running:

1) Asterisk PBX since summer 2014 (probably should replace the SD card but it really juts boots it most stuff runs from the USB drive)
2) 2nd DB server for door control since about the same time probably later summer 2014. Could be off for all the use it gets, but rules say you need 2 DB servers.
3) The Pi-Hole & open VPN server was up from 2015 until I got my 1st RPi4 and wanted to play with it. It's actually the one running now as I wipe / re-create the one I am asking about
4) My torrent server on a RPi 3 from winter 2016/17


Agreed, when diagnosing issues with RPi it's usually this or an under powered power supply.
It's amazing how many people spend real money on cases and other things for their stuff and then go for the $1.99 power supply and SD card.

-Dave

my experience is only with a 3B+, and I ended up downclocking/undervolting the CPU to get it stable. Things like that and SD card wear make the standard Pi as delivered not suitable for server type applications, although....



.....that's a good track record! As I mentioned though, anything that makes much use of the SD card (i.e. swap, logging) will put your uptime into question. So using the SD just for boot (as you do for your PBX server) is a smart move.

As you know new bitcoin version is out.
In the Release Notes: Bitcoin Core version 0.19.0.1 (https://github.com/bitcoin/bitcoin/blob/master/doc/release-notes/release-notes-0.19.0.1.md), I can read:


This might be worth noting when running bitcoind on a Raspi4.

yep

0.19 also adds 2 extra outbound connections to protect against a certain class of network-level attack. That means you're using slightly more memory than version 0.18 and before. So the 9% reduction in overall memory usage offsets that change nicely. Still a good idea to increase dbcache by at least 5% I would say, even if you're using the 1GB variant of RPi4 (and when you're syncing the chain, definitely use as much dbcache as you can temporarily)

I had some issues with the WIFI for a while with my RP4 and googled a bit for the problem. Surprisingly what I found is that HDMI is jamming it's own WIFI at 2560×1440 resolution, and I'm not the only one to have it, so this is something you should be aware of.

Here's the source and interesting read, short but interesting. > https://hackaday.com/2019/11/28/raspberry-pi-4-hdmi-is-jamming-its-own-wifi/
Anyone having issues with the WIFI?

Interesting, I use mine as a IPTV box now and I thought that the issues are  because of the heating problem but It turns out that is this jamming issue. After reading this artichel  I changed the resolution to 1080p it goes good. Try to see if this will help with your case as well it might be the same situation. I hope they find a solution.
 

5GHz Wifi has shorter useful range, I would consider just connecting the Pi4 over ethernet cable (and removing the Wifi driver module from the kernel build)

well, just that it sounds like an unusable piece of hardware (i.e. the WiFi chip when you're using HDMI displays of certain resolutions), so you may as well remove the driver module in case some other software on the Pi4 tries using the WiFi connection in future. In general, removing modules from the kernel you don't have a use for is a good idea, you can just compile it again should there be some module you later want to use

I am fairly anti rasp pi 🕵️‍♀️.

They are underpowered and can’t quite get the job done.

there are so many used lenovo tiny pcs available that simply are 10 to 100x better then a rasp pi.

I don’t understand why try to push a rasp pi past what it can do.

especially if it is wallet involved.  A lenovo m700 tiny

used on ebay under 150.

  it can use both a m2 ssd and a 2.5 ssd and have 16gb ram.  does wifi well.

they have (arguably) at least 2 things going for them that Intel/AMD does not:

• reasonably programmable firmware
• cheap

in the case of this Wifi/HDMI crosstalk issue on the Pi4, the latter point is actually a downside. RPi product development has always been driven by low price tags, and that's also always been reflected in the reliability/performance.

But overlooking the first point is also unwise, there's not much point in an independent bottom-up currency being run on hardware with potential corporate backdoors (basically a rootkit operating way below the level of the owner's root access, and that can be accessed over the internet, even when powered over ethernet while the main power supply is switched off :-\). If the whole Bitcoin network was running on such hardware, then Intel/AMD could in essence be in possession of a Bitcoin off-switch.

I have owned each and every rasp pi other then the 4.

they all feel and act so underpowered compared to a lenovo m tiny or a mac min or a dell micro or an hp elite.

i run a lot of those four units from four companies.

but they all have intels in them.

even though i use linux win 7 win 10 and mac os.

I have yet to get a rasp pi to feel powerful enough to run like the intel models. but i guess having a third set of cpus running a core is not a bad idea.

maybe i will spring for a rasp pi. 4

yeah, Raspi's have always had underpowered CPUs. The RPi4 is a step forward, but really because of the 2/4GB RAM variants. CPU is better, but not alot. I think it has more crypto instructions in the ISA (Cortex A72 I think), but not SHA256 sadly. Supposedly SHA256 acceleration only improves Bitcoin tx validation performance on x86_64 by 5%, but maybe the improvement would be bigger on a future RPi, as it would be improving from a lower performance baseline.

For those wondering how long it takes to sync,


549sec for 136 block

4 seconds per block

622262*4/60/60/24 = 28.8 days, assuming full block + segwit.

Yeah, that seems a bit slow. If you are on the high end of hardware.
RPi4 / 4GB / Samsung USB SSD / 300MB connection (before corona and everyone working from home killed my connection)  did if from scratch in under 10 days.
I keep feeling that towards the end of the sync when most block are full a spinning drive just kills performance vs. an SSD

-Dave