EKOS debugger shows this error:

I found indi-libcamera was missing, installed it, no change.
Ideas please?

Read More...

Also when using simulators drivers and starting up ekos, defining a job, running it shows:

Connection to INDI host at localhost on port 7,624 lost. Server disconnected: -1
after a few seconds of job start (when status would be switching to "slewing")

Then, when re-starting ekos, i would be warned that another indi server is running, to which i can't connect.
so starting a new indi server gets me to the old problem with job start. sends me into the error->idling loop, with no debug output whatsoever.

Read More...

Updated SM to the latest version (upgrade and dist-upgrade).
After defining and running a job i get an error message: "Connection to localhost:***** failed", job status cycling between Idle, Slewing and Error.
When restarting the INDI server via EKOS, the following appears in the log, repeated every few seconds:

[2023-06-17T22:52:57.766 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Scheduler is awake."
[2023-06-17T22:52:57.770 CEST INFO ][ org.kde.kstars.ekos.scheduler] - Scheduler is starting...
[2023-06-17T22:52:57.771 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Scheduler started."
[2023-06-17T22:52:58.828 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'IC 1396' 60x120\"  sees 0 captures in output folder '/home/stellarmate/Pictures/IC_1396/Light'."
[2023-06-17T22:52:58.828 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'IC 1396' 60x120\"  requires a dither procedure."
[2023-06-17T22:52:58.828 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'IC 1396' estimated to take 02h 18m 45s to complete."
[2023-06-17T22:52:58.856 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "IC 1396   \t06/17  22:52 --> 01:11 \tjob completion"
[2023-06-17T22:52:58.858 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Greedy Scheduler plan for the next 48 hours starting Sat Jun 17 22:52:58 2023 (0.03)s:"
[2023-06-17T22:53:00.966 CEST INFO ][ org.kde.kstars.ekos.scheduler] - Executing Job  "IC 1396"
[2023-06-17T22:53:01.970 CEST CRIT ][ org.kde.kstars.ekos.scheduler] - "Warning: mount canPark request received DBUS error: 20"
[2023-06-17T22:53:01.971 CEST CRIT ][ org.kde.kstars.ekos.scheduler] - "Warning: job 'IC 1396' slew request received DBUS error: org.freedesktop.DBus.Error.UnknownObject"
[2023-06-17T22:53:02.976 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'IC 1396' is stopping guiding..."
[2023-06-17T22:53:03.976 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'IC 1396' 60x120\"  sees 0 captures in output folder '/home/stellarmate/Pictures/IC_1396/Light'."
[2023-06-17T22:53:03.976 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'IC 1396' 60x120\"  requires a dither procedure."
[2023-06-17T22:53:03.977 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'IC 1396' estimated to take 02h 18m 45s to complete."
[2023-06-17T22:53:04.038 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "IC 1396   \t06/17  22:53 --> 01:11 \tjob completion"
[2023-06-17T22:53:04.042 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Greedy Scheduler plan for the next 48 hours starting Sat Jun 17 22:53:03 2023 (0.068)s:"
[2023-06-17T22:53:05.004 CEST INFO ][ org.kde.kstars.ekos.scheduler] - Executing Job  "IC 1396"
[2023-06-17T22:53:05.510 CEST INFO ][ org.kde.kstars.ekos.scheduler] - Scheduler is stopping...
[2023-06-17T22:53:05.512 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'IC 1396' is stopping guiding..."
[2023-06-17T22:53:05.513 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'IC 1396' has not been processed upon scheduler stop, marking aborted."

I am running ASI120Mini via PHD2, ASI294MC Pro on a EQ5 mount. Everything was running smooth when i started it up last time (three months ago).

Any tips, please?

Thanks,
MM

Read More...

Well, you can actually clone a linux installation to a thumb drive and let it run off it (if your target computer supports booting from usb storage devices).
In this case it would be pointless, because you need a installed and preconfigured Stellarmate X OS as well as a Linux live cd with some tools for preparing the thumb drive as an actual harddrive (with ext3/4 and vFAT partitions) and clone the unmounted drive contents to it.
It would be way easier and more convenient to use by just installing on an internatl SSD.

Read More...

Hi,

there is an easy but effective remedy: An iPad stylus :)
I was fat-fingering almost everything on my iPad Mini, which was too old and weak for the SM app, too.
When i replaced it for a used Samsung S6 Lite Tab, which included a stylus, i even almost enjoyed doing occasional VNC sessions with the Stellarmate Pi. It's also crucial to set a useful desktop resolution for the remote session.

Read More...

Mr. Mister replied to the topic 'No discount for OS X?' in the forum. 12 months ago

Dual Boot:
If the Beelink lets you choose the boot drive on startup (BIOS level), you could flash SMX onto a external SSD and connect it via USB3.
I am running classic SM on such a configuration, but in my case the boot partition is on the SD card, because the Pi doesn't have a boot selector.

Read More...

Imo, flashing to a thumbdrive would work fine if your (PC) hardware is compatible with the kernel and modules of the SMX distribution (OS) and your PC is able to boot from USB. At some point some config editing would be required, so the easiest route would be unplugging your system drive, installing a fresh low price SATA SSD (or old hard drive) and install SMX there to try.
Just plug back the windows drive and it will run as usual.

I plan to upgrade to SMX, maybe also the hardware, as i'm a happy SM user on RasPi hardware, as it's possible to control my stuff with the app (on the road) or via RDP (or VNC) directly on the Pi in my home network. No other product is offering this to me.

Read More...

Mr. Mister replied to the topic 'Issue with QHY driver' in the forum. 1 year ago

As far as it comes to Linux, the compiler/maker references libraries it can find through environment variables ($LD_LIBRARY_PATH, $PATH etc.) set globally or locally (in the script, possibly wrapping script or through the configure run) and in /usr/local/lib/, /usr/lib/ (...) most .so "files" are sysmbolic or hard links to the current versions of installed libraries beneath or in this directory.

For example /usr/share/lib/libld.so would be a link to /usr/share/lib/libld.so.5, so the compiler(s) are using the symlink to access the linked shared library file.

This means you don't have to delete library files, but change the links (look at the "ln" command) to point to the compatible library, if no direct reference was given in ./configure or passed to the maker/compiler through environment variables, the compiler will use the links to find the desired libraries. Normally you would not delete the newer libraries in /usr/lib in this case, but let the links point to the .so files having the right version number in their names.

Doing it in any other way would be quite a mess, it's done through links for ages in Unix and Linux, for good reasons.
Maybe Raspi Linux is an exception, i don't have a Raspi nearby at the moment to check, but i think basic things are done the same way here.

Read More...

I also have experience with tar and zlib, which is best suited for backups of data of all sorts, but it's just not really convenient to backup whole partitions/disks, because it doesn't respect device files, as well as mountpoints or mounted partitions. But if you want to backup user data, like fits/raw image files, this is the go-to command. It can also work incrementally, so you don't need much space for more than one full backup on your backup space/medium.

When i was dealing with Unix in enterprise environments, i used a mix of dd images of base installs and incremental/differential tar archives, which got zipped on the fly. I also upgraded my Pi4 with a SDD, which sped up Stellarmate significantly. This, however, adds the burden to do some extra filecopy work in the process of freshly installing Stellarmate OS from official sd-crad image files, but on the upside, backups of the SSD (root partition) are lightning fast, only the boot-partition is still needed on the sd-card for the OS, to know to mount the root partition from the SSD.

Please post back your findings of using dd with the "sparse" option set, thanks :)

Read More...

Sara, this is a great advice for shrinking backups after they're created, but this thread is about backup time at image creation.

There's another possibility, by using the "sparse" option of dd:

serverfault.com/questions/439128/dd-on-e...t-want-empty-portion

However, i didn't test it (yet), but seems worth a try if you have some time to enter commands and let the system do the work.
Append calls of "date" pre and post execution like so:

date && someothercommand && date

It will echo start and finish date/time before and after someothercommand is being executed. So you can leave the screen while dd is running.
Writing back the dd image to sdcard would be best with omitting the "sparse" option, i guess.

Read More...

Mr. Mister replied to the topic 'Issue with QHY driver' in the forum. 1 year ago

Hi, glad you got it resolved.
Please post a link to the script, for other users looking to solve the same problem(s).

Clear skies!

Read More...

Mr. Mister replied to the topic 'Issue with QHY driver' in the forum. 1 year ago

Max i redirect you to this thread, please?

www.indilib.org/forum/ekos/5392-kstars-c...raspbian-buster.html

In essence: Install all prerequisite software and libraries, also the optional ones & use cmake.

Read More...