Official FutureBit Apollo LTC Image and Support thread

[WaterTower]

Can we run these in turbo mode yet? Has anyone tried?

I tried turbo mode in the initial release to just play around with it, and it seemed to work fine for the short time I tested it. The only issue was that I only felt comfortable with the fan at 100%, and while it will keep the unit cool enough, the sound it makes does not give me a lot of confidence that the fan will last a long time at that speed. I keep mine in balanced mode with fan manually set to 20%.

[michelem]

Downloading from link provided, 500 KB/s... (I have 75 Mbps Fibre if you ask)
Is there chance for torrent or something? Torrent could be even hosted in Apollo itself and distributed to people. So no need for AWS.

Did you try the wetransfer link? Ill see if I can put it up on mega or something as well.

jstefanop I just added the image to my server placed in Germany. If someone needs it this is a mirror: http://getminera.com/releases/apollo_image_1-29.img.zip

[VictorGT]

I would like to put static IP adresses to my Apollos. Is this possible with latest software?
Also, can we test BALANCED and TURBO modes now?

Ive added it to my list

Great... 

[whiteogre]

I would like to put static IP adresses to my Apollos. Is this possible with latest software?

That configuration needs to still be done from the command line, like in the previous release, if you want to do it from the Apollo side. Usually it's more practical to configure the dhcp server to provide some specific IP for the matching MAC address of each Apollo. That way, the IP is always the same regardless of software updates and no command line access is needed.

[jstefanop]

-Download the new Apollo image linked above and make sure you UNZIP it (always a good idea to check MD5 checksum in case download was corrupted...make sure checksum is run on the .img file NOT the downloaded .img.zip)
-Use Etcher to re-image your SD card with the updated image file

Etcher has supported already for few years extracting the image from the zip on the fly as long as the zip contains only one file, like is the case here. Therefore, there shouldn't be any real reason to unzip first. I also think you should take the checksums from the zip file itself since that's essentially the release you are distributing. At least I see it so that the checksum can be used to validate that the release is indeed what it should be and not something that has been replaced with a botnet addon. That's why the checksum should be of the distributed file instead of a file within it. Zip by itself will anyway create an internal checksum of the file and extracting it will fail if the file has been somehow corrupted during transfer. If you'd want to go the pro way then signing the release file with gpg would be the way to go. 

Few comments regarding the release itself, less than previously thanks to fixes:

• the login screen password input field still doesn't grab keyboard focus automatically on page load
• serial-getty@ttyS0.service is still active and constantly trying to start and failing to do so
• /var/log.hdd could use a cleanup before release
• documentation for front led blink signals is missing from the first post, I've seen at least slow yellow, fast yellow, slow green and irregular red

Yea this release was a bit rushed since I wanted to get the hardware watchdog implemented ASAP. Everything else is on the list for the next release.

As for documentation I am still working on a really thorough "User Guide" that is taking way longer than I like to write up in between all the software updates, and PSU I'm working on. Im forcing myself to finish that up next week.

As for LEDs here is the basic description:

-Slow Blinking Yellow LED: System Boot/Hardware initialization
-Fast Blinking Yellow LED: Hardware passed all checks/inits, started bfgminer/connecting to pool
-Slow Blinking Green LED: Bfgminer successfully started and connected to pool
-Slow Blinking Green LED + Random Red flashes: Normal miner operation...Red flashes indicate each share returned by ASIC board and show low level miner activity
-Solid Red LED: Hardware fault or system shutdown (Red LED during normal operation means hardware monitor detected a fault with either the voltage regulator, Fan operation, or abnormal PCB temperatures).

[jstefanop]

Can we run these in turbo mode yet? Has anyone tried?

I tried turbo mode in the initial release to just play around with it, and it seemed to work fine for the short time I tested it. The only issue was that I only felt comfortable with the fan at 100%, and while it will keep the unit cool enough, the sound it makes does not give me a lot of confidence that the fan will last a long time at that speed. I keep mine in balanced mode with fan manually set to 20%.

Yes Turbo mode is safe at this point. This release has additional layers of protection that will prevent any nuclear meltdown scenarios in turbo mode 

I would not mess with custom mode until I have the full guide up for this (if you have plenty of GPU/ASIC experience with voltage controls and frequency tuning that's fine as long as you don't do anything crazy)

[philipma1957]

Can we run these in turbo mode yet? Has anyone tried?

I tried turbo mode in the initial release to just play around with it, and it seemed to work fine for the short time I tested it. The only issue was that I only felt comfortable with the fan at 100%, and while it will keep the unit cool enough, the sound it makes does not give me a lot of confidence that the fan will last a long time at that speed. I keep mine in balanced mode with fan manually set to 20%.

Yes Turbo mode is safe at this point. This release has additional layers of protection that will prevent any nuclear meltdown scenarios in turbo mode 

I would not mess with custom mode until I have the full guide up for this (if you have plenty of GPU/ASIC experience with voltage controls and frequency tuning that's fine as long as you don't do anything crazy)

Mid speed with fans on manual at 25% work well

121 mh  and temps are 60c  45c

[jstefanop]

I would like to put static IP adresses to my Apollos. Is this possible with latest software?

That configuration needs to still be done from the command line, like in the previous release, if you want to do it from the Apollo side. Usually it's more practical to configure the dhcp server to provide some specific IP for the matching MAC address of each Apollo. That way, the IP is always the same regardless of software updates and no command line access is needed.

Yea, keep in mind I know most of you here are experienced miners, but the goal of this platform is to onboard every day consumers into crypto with an easy to use interface so I dont want to clutter/overwhelm new users with a large number of options (Will probably move a lot of the more advanced settings into a new "Advanced Section" either way though).

Obviously want to allow advanced options for you guys as well, but stuff like this that is easily done outside the platform is low on priorities (for example MAC addresses are displayed in the system info pane, and almost all routers allow you to assign a specific MAC to an IP address as mentioned above).

[VictorGT]

Thanks. I know about DHCP/MAC/ROUTER settings. No problem. A little lazy  

[crypto_curious]

Downloading from link provided, 500 KB/s... (I have 75 Mbps Fibre if you ask)
Is there chance for torrent or something? Torrent could be even hosted in Apollo itself and distributed to people. So no need for AWS.

I could download it just right now with 25MB/s in less than a minute from home.
You must be holding it wrong   (or your providers pipe to amazon is a tad slow).

Just tried the download again, speed is 500-600 KB/s peak.

I live in England, and this 75Mbps Fibre is perfect in everyday use, I don't get any hiccups or slowdowns, and I have server here, which is 24/7 online, registers no downtimes or anything for months. Take a look at trace route:

Code:

>tracert s3.amazonaws.com

Tracing route to s3-1.amazonaws.com [52.216.226.123]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  vodafone.broadband [192.168.1.1]
  2    13 ms    13 ms    13 ms  host-212-158-250-33.dslgb.com [212.158.250.33]
  3    15 ms    15 ms    15 ms  63.130.127.221
  4    15 ms    15 ms    15 ms  ae15-xcr1.slo.cw.net [195.2.2.217]
  5    82 ms    83 ms    83 ms  ae2-xcr1.nyh.cw.net [195.2.28.169]
  6    82 ms    82 ms    82 ms  ae13-xcr2.nyk.cw.net [195.2.25.69]
  7    82 ms    82 ms    82 ms  52.95.216.78
  8    95 ms   123 ms    97 ms  52.93.4.75
  9    88 ms    88 ms    90 ms  52.93.4.36
 10     *        *        *     Request timed out.
 11    88 ms    88 ms    88 ms  54.240.229.137
 12     *        *        *     Request timed out.
 13     *        *        *     Request timed out.
 14     *        *        *     Request timed out.
 15     *        *        *     Request timed out.
 16     *        *        *     Request timed out.
 17     *        *        *     Request timed out.
 18     *        *        *     Request timed out.
 19    89 ms    88 ms    89 ms  52.93.28.140
 20     *        *        *     Request timed out.
 21     *        *        *     Request timed out.
 22     *        *        *     Request timed out.
 23     *        *        *     Request timed out.
 24     *        *        *     Request timed out.
 25     *        *        *     Request timed out.
 26    89 ms    88 ms    88 ms  s3-1.amazonaws.com [52.216.226.123]

Trace complete.

After 7th hop, I am already in Amazon network. And I stay there until 26th hop. All these IPs 52.95.x.x, 52.240.x.x and 52.93.x.x belongs to Amazon. My ISP handles all traffic to Amazon and there is gets tangled up. I am not sure what "request timed out" mean in this case, though.

It's not that it's a big deal, but someone is paying for services, which don't get delivered Either my ISP or AWS is not doing their thing.

[Globalminingcorp]

Quick download from wetransfer and update process itself took about 10 minutes. The MCU stays a lot cooler than before (about 10 degrees celsius). I tried to lower the fan to 5% but than the miner temperature got into 75+ degress so i went back to my 'old' 10% fanspeed and miner temperature got to 'normal' 62ish.

Thanks for the update, looking forward to more.

[jstefanop]

Downloading from link provided, 500 KB/s... (I have 75 Mbps Fibre if you ask)
Is there chance for torrent or something? Torrent could be even hosted in Apollo itself and distributed to people. So no need for AWS.

I could download it just right now with 25MB/s in less than a minute from home.
You must be holding it wrong   (or your providers pipe to amazon is a tad slow).

Just tried the download again, speed is 500-600 KB/s peak.

I live in England, and this 75Mbps Fibre is perfect in everyday use, I don't get any hiccups or slowdowns, and I have server here, which is 24/7 online, registers no downtimes or anything for months. Take a look at trace route:


After 7th hop, I am already in Amazon network. And I stay there until 26th hop. All these IPs 52.95.x.x, 52.240.x.x and 52.93.x.x belongs to Amazon. My ISP handles all traffic to Amazon and there is gets tangled up. I am not sure what "request timed out" mean in this case, though.

It's not that it's a big deal, but someone is paying for services, which don't get delivered Either my ISP or AWS is not doing their thing.

Its hosted on an east coast AWS server so could be that amazon does not have optimized routes to europe. Ill see about getting a European bucket up.

[whiteogre]

There could be a small issue with the latest release. The added watchdog appears to be a little bit chatty causing the syslog to be filled with

Code:

watchdog: watchdog0: watchdog did not stop!

messages at a rate of about 12 every minute. Normally that wouldn't be an issue, but the /var/log mountpoint on the Apollo is ram disk sized to 50MB regardless of the used sd card size. It's filled at around 53% after install and there doesn't appear to be any cleanup or logrotate for this mountpoint active. As a result, the mountpoint will be full in around 2 weeks at the current rate. Most likely not all system services will be happy once no more logs can be written.

[VictorGT]

I call my miners Apollo1 y Apollo2 (not too much original names hum...  ).  I put a couple of quality labels on each (on top and front).

But, no way to know what miner I am monitoring when using webgui... only by IP.

I will like a lot a simple text field for assigning each miner a name. If visible at dashboard, you can identify your miners very easy.

Maybe too stupid request, but really easy to implement ha ha...

[crypto_curious]

I call my miners Apollo1 y Apollo2 (not too much original names hum...  ).  I put a couple of quality labels on each (on top and front).

But, no way to know what miner I am monitoring when using webgui... only by IP.

I will like a lot a simple text field for assigning each miner a name. If visible at dashboard, you can identify your miners very easy.

Maybe too stupid request, but really easy to implement ha ha...

You can recognize your miners by worker in your pool, visible on the same dashboard page you mentioned. Just open dashboard page, scroll down, there will be your pool address and worker name. Name them accordingly - youraccount.Apollo1 and youraccount.Apollo2 workers. Job done.

[VictorGT]

I call my miners Apollo1 y Apollo2 (not too much original names hum...  ).  I put a couple of quality labels on each (on top and front).

But, no way to know what miner I am monitoring when using webgui... only by IP.

I will like a lot a simple text field for assigning each miner a name. If visible at dashboard, you can identify your miners very easy.

Maybe too stupid request, but really easy to implement ha ha...

You can recognize your miners by worker in your pool, visible on the same dashboard page you mentioned. Just open dashboard page, scroll down, there will be your pool address and worker name. Name them accordingly - youraccount.Apollo1 and youraccount.Apollo2 workers. Job done.

Sorry. I have not any worker name field in webgui. Sometimes I have that info at Username field (but not always). My pool address do not show any worker name in Dashboard.

Screenshot: https://www.dropbox.com/s/zl3y6qb2mb1htcx/noworkername.png?dl=0

How can i know my worker name?

I can see my workername with Litecoinpool AT USERNAME field, but not all pools works the same way.

EDIT: prohashing uses password field to put workername, and that field, of course, is not visible at Dashboard.

[crypto_curious]

Oh, you screwed then If prohashing uses password field to determine worker, then it's very strange behaviour. Yeah, I get it, you are not able to determine which Apollo you looking for just by looking at Apollos dashboard. They would be all the same to you.

I, for a change, use pool where my login is username.workername, I can determine which Apollo I'm looking at just by looking at the dashboard because my username there will be all I need to see.

Another suggestion for you then:
Setup up your router, so MAC address or Apollo is assigned to static IP. That way, they have same IPs forever.

[VictorGT]

Oh, you screwed then If prohashing uses password field to determine worker, then it's very strange behaviour. Yeah, I get it, you are not able to determine which Apollo you looking for just by looking at Apollos dashboard. They would be all the same to you.

I, for a change, use pool where my login is username.workername, I can determine which Apollo I'm looking at just by looking at the dashboard because my username there will be all I need to see.

Another suggestion for you then:
Setup up your router, so MAC address or Apollo is assigned to static IP. That way, they have same IPs forever.

Anyway, not a big problem. I can go to pools page and read Password field.

[whiteogre]

I call my miners Apollo1 y Apollo2 (not too much original names hum...  ).  I put a couple of quality labels on each (on top and front).

But, no way to know what miner I am monitoring when using webgui... only by IP.

I will like a lot a simple text field for assigning each miner a name. If visible at dashboard, you can identify your miners very easy.

Maybe too stupid request, but really easy to implement ha ha...

This falls to the category of "do only if you know you can copypaste and follow the instructions" but here's how you can change the page title of the dashboard of each unit. The title is what's visible in the browser window bar or tab bar, not on the page itself. You need to connect via ssh to the mcu (the login and password has been mentioned in earlier posts) and then run the following two commands. Edit only the first one to match the wanted title but don't touch the second line:

Code:

pagetitle="The new dashboard title comes here within these quotes"
sed -i -e "s#<title>.*</title>#<title>${pagetitle}</title>#g" /opt/apolloapi/build/index.html

Reload the dashboard in the browser and that should be it. The original title is "Apollo miner web dashboard" if you forgot to write it down.

[jstefanop]

There could be a small issue with the latest release. The added watchdog appears to be a little bit chatty causing the syslog to be filled with

Code:

watchdog: watchdog0: watchdog did not stop!

messages at a rate of about 12 every minute. Normally that wouldn't be an issue, but the /var/log mountpoint on the Apollo is ram disk sized to 50MB regardless of the used sd card size. It's filled at around 53% after install and there doesn't appear to be any cleanup or logrotate for this mountpoint active. As a result, the mountpoint will be full in around 2 weeks at the current rate. Most likely not all system services will be happy once no more logs can be written.

Yea this is hardcoded in the kernel source and didn't have time to remove it. Even if the log fills its not a huge deal but will have a process to clean up the logs next release.