SDS800X HD Bug Reports + Firmware

[newbrain]

Latest FW, 1.1.6.2, tested saving various formats on:

• TrueNAS Core, TrueNAS-13.0-U6.1, samba 4.15.13, using a domain user EtA: or a local NAS user,  share has >3 TB free:
  Can only save screenshots (bmp, png, jpg), reference trace (ref) and setup (xml), anything else is met with "Not enough storage available!" message
• TrueNAS Core, TrueNAS-13.0-U6.1, samba 4.15.13, but using a share with guest access, with guest, or a domain user, or a local NAS user, share has 35 GB free:
  All file types can be saved.
• A shared folder on Linux Mint 22.1, using samba 4.19.5-Ubuntu, guest access:
  I'm able to save any type of file with no issue.
• Shared folder on a Windows 10 Pro 22H2 machine, domain user:
  No problems with any format.
• Shared folder on a Windows 11 Pro 23H2 machine not in the domain, local Windows user:
  No problems with any format.

Yes, Windows machines have oldish versions, I mostly use Linux.
The procedure was always the same:
Utility, System, Services, SMB client, Connect the client using the machine's IP address and filling in the user credentials (apart from the guest).
Utility, Save/Recall, select file type, File Manager, Save as, repeated for the various types.

It seems that the problem appears (at least in my testing) when using a domain user on samba shares - but not on Windows shares.

Nope! See the added red text!
I suspect the free space plays a role: the samba share I used for the first test had more than 3 TB available, alla the others, less than 300 GB.

Small possible bug on keyboard handling from the web interface: I'm using a Swedish keyboard, many shifted keys do not match, e.g. Shift-7, which should give a forward slash '/', generates a question mark on the scope, Shift-, (comma), instead of giving a semicolon, generates a colon. No differences between Linux and Windows, Chromium or Firefox. US layout works correctly.

[Furna]

I suspect the free space plays a role: the samba share I used for the first test had more than 3 TB available, alla the others, less than 300 GB.

I am not sure ... take a network packet capture of a PNG and a REF; see the different behaviour.

[newbrain]

I am not sure ...

I wasn't 100% confident either.

But I am now.
I have created two identical samba shares on the TrueNAS server (irrelevant: I have in the meantime updated it to Scale), with the only difference being that one had a 100 GB quota, the other 3 TB.

Saving of bin etc. fails only on the larger one, with the usual not enough storage message.



So I would definitely say that the discriminant factor is the available space, if I weren't lazy I might try to bisect it.
Looks like I'm not that lazy after all...
To be honest I did not bisect, I went directly for the kill.

The threshold is 2³¹ bytes - typical use of a signed 32 bit integer when something quite larger would be needed? - few bytes less and it's OK, few bytes more and you get the error message.

@2N3055, @tautech, this might be worth reporting?

[2N3055]

.............................
@2N3055, @tautech, this might be worth reporting?

Absolutely..
Good job.

Will report.

[Furna]

if I weren't lazy I might try to bisect it.
Looks like I'm not that lazy after all...
To be honest I did not bisect, I went directly for the kill.

Let me the spot for number one lazy person around
Well done!

.............................
@2N3055, @tautech, this might be worth reporting?

Will report.

Thank you

[newbrain]

Will report.

Thanks!

Can you also verify the keyboard problem with the web server?
I also tried the Italian and the Croatian layouts in addition to the Swedish one - they are quite similar for the involved keys - with the same results.
Shift-7 -> ? instead of /
Shift -, -> : instead of :
Some mix-up also with + and =.

[mikes]

The threshold is 2³¹ bytes - typical use of a signed 32 bit integer when something quite larger would be needed? - few bytes less and it's OK, few bytes more and you get the error message.

So strange. What could they be doing, that it only affects some file types? Not checking free space before saving for the ones which work? That would be wrong, too.

[AndyC_772]

Quite possibly reporting "not enough space" as a default any time an attempt to create a file doesn't work.

After all, what else could ever go wrong? 

[newbrain]

What could they be doing, that it only affects some file types?

Maybe they have slightly different ways to check for enough free space?

Where in the working one they use 64 bit integers, and in the wrong one 32 bit, or some such sloppy mixing of int sizes and signedness...
https://godbolt.org/z/9oreTTahz

[tszaboo]

I was doing some I2C decoding. I2C scanning with the microcontroller. Some device wasn't responding to it's address.
Please see below. Somehow the decoding stops working after address 0x77. Which is supposedly a the address of the device that's not working. So why am I not getting any decoding? And if the decoding still works, why am I not getting any frame error or any other error indications?

[tautech]

I was doing some I2C decoding. I2C scanning with the microcontroller. Some device wasn't responding to it's address.
Please see below. Somehow the decoding stops working after address 0x77. Which is supposedly a the address of the device that's not working. So why am I not getting any decoding? And if the decoding still works, why am I not getting any frame error or any other error indications?

Please confirm firmware version installed.

[tszaboo]

I was doing some I2C decoding. I2C scanning with the microcontroller. Some device wasn't responding to it's address.
Please see below. Somehow the decoding stops working after address 0x77. Which is supposedly a the address of the device that's not working. So why am I not getting any decoding? And if the decoding still works, why am I not getting any frame error or any other error indications?

Please confirm firmware version installed.

1.1.3.1 with HW version 00.

[tautech]

I was doing some I2C decoding. I2C scanning with the microcontroller. Some device wasn't responding to it's address.
Please see below. Somehow the decoding stops working after address 0x77. Which is supposedly a the address of the device that's not working. So why am I not getting any decoding? And if the decoding still works, why am I not getting any frame error or any other error indications?

Please confirm firmware version installed.

1.1.3.1 with HW version 00.

You're at least a couple of versions behind current.....and we are already testing the next version.
https://int.siglent.com/download/firmwares/?ProId=74

[tszaboo]

I was doing some I2C decoding. I2C scanning with the microcontroller. Some device wasn't responding to it's address.
Please see below. Somehow the decoding stops working after address 0x77. Which is supposedly a the address of the device that's not working. So why am I not getting any decoding? And if the decoding still works, why am I not getting any frame error or any other error indications?

Please confirm firmware version installed.

1.1.3.1 with HW version 00.

You're at least a couple of versions behind current.....and we are already testing the next version.
https://int.siglent.com/download/firmwares/?ProId=74

Same thing happening with the latest FW I could download from here:
https://www.siglenteu.com/service-and-support/firmware-software/digital-oscilloscopes/#sds800x-hd-series

Your link didn't work.

[starzgames]

Hi guys, just saw there was an updated firmware. Scope had *.36 version, went to do the firmware update, it rebooted, relays clicked a few times, all the lights turned on, and screen just says SIGLENT for the last 30 minutes. I used a freshly formatted flash drive with only the updated firmware file "SDS800X-HD_1.1.6.2.ADS" on the root of the flash drive, with nothing else on it.

It asked me 'reboot now?', I clicked yes, it rebooted, and now it is just hung. I've tried connecting to it over ethernet and over usb to see if there is any life, but there isn't any. I haven't touched it / turned it off, but I'm pretty sure I just bricked the fucking thing.

Edit: I couldn't take it anymore. I pressed and held power button for 15 seconds before it turned off. Unplugged everything, and turned it back on. sat there for several lifetimes, then went to the normal screen. Checking system information shows the bios did upgrade correctly.

I have no idea what happened, but I cannot put into words how relieved I am. Except I just did reset defaults to make sure, clicked reboot, and now it's stuck on the black screen again. Motherfucking what.

Final edit: I found the problem. I have a usb powered wifi travel router. It plugs into the back usb/ethernet ports, and connects to my wifi automatically. Somehow I think on bootup, it seeing the router maybe did a programmed thing like for a tech to flash a firmware of a dead device over tftp or flash etc.

Unplugging it, it turns on jus fine, plug it in, all is well. It just messes with bootup somehow

[2N3055]

Hi guys, just saw there was an updated firmware. Scope had *.36 version, went to do the firmware update, it rebooted, relays clicked a few times, all the lights turned on, and screen just says SIGLENT for the last 30 minutes. I used a freshly formatted flash drive with only the updated firmware file "SDS800X-HD_1.1.6.2.ADS" on the root of the flash drive, with nothing else on it.

It asked me 'reboot now?', I clicked yes, it rebooted, and now it is just hung. I've tried connecting to it over ethernet and over usb to see if there is any life, but there isn't any. I haven't touched it / turned it off, but I'm pretty sure I just bricked the fucking thing.

Edit: I couldn't take it anymore. I pressed and held power button for 15 seconds before it turned off. Unplugged everything, and turned it back on. sat there for several lifetimes, then went to the normal screen. Checking system information shows the bios did upgrade correctly.

I have no idea what happened, but I cannot put into words how relieved I am. Except I just did reset defaults to make sure, clicked reboot, and now it's stuck on the black screen again. Motherfucking what

Interesting.

How long did you wait to reboot scope after loading defaults? Immediately after few seconds or after few minutes?

Also do you have Ethernet cable connected? Do you use DHCP?

[starzgames]

Hi sorry, just edited my post. Everything is working just fine, I just reset defaults after it booted up again. Somehow the ethernet port connected to the travel router interferes with bootup, and causes it to hang/freeze

[2N3055]

Hi sorry, just edited my post. Everything is working just fine, I just reset defaults after it booted up again. Somehow the ethernet port connected to the travel router interferes with bootup, and causes it to hang/freeze

Yes that is why I asked about Ethernet. There is a small confirmed bug in last firmware (they are working on a fix) that if you are using DHCP, and Ethernet Cable is plugged in, if you cannot reach DHCP server on boot it can get stuck on Start screen. To avoid problems until fix is published, either use fixed IP or make sure DHCP server is ready, or simply connect cable after boot.
And no, I don't know when exactly fix will be published, sorry.

[starzgames]

So yeah, I took years of my life off on that one. Wish I had known that tidbit before thinking firmware update failed and I bricked my brand new meter!

[2N3055]

So yeah, I took years of my life off on that one. Wish I had known that tidbit before thinking firmware update failed and I bricked my brand new meter!

Yes, I  can imagine it was not fun....

[tautech]

So yeah, I took years of my life off on that one. Wish I had known that tidbit before thinking firmware update failed and I bricked my brand new meter!

Fear not, Siglent has many tools to recover instruments if/as required.

As 2N3055 explains, this is a reported bug with the current FW but I haven't seen this boot hang with it or with the beta version we have now.

[Furna]

[...] There is a small confirmed bug in last firmware (they are working on a fix) that if you are using DHCP, and Ethernet Cable is plugged in, if you cannot reach DHCP server on boot it can get stuck on Start screen[...]

Where is this reported? I hit this bug a few days ago while installing a new switch dedicated to the test devices and interconnection to "main" switch was not working due to a faulty cable; as such DHCP was not reachable.
To reproduce anyway I had to di completely disconnect the power to the scope.
I was going to report here (I have been busy inthe last few days) when I read 2N3055.
Nevertheless ... happy to learn it has been already signaled to Siglent but I lost the relevant post here in this thread. Last message aboot this I am aware of is https://www.eevblog.com/forum/testgear/sds800x-hd-bug-reports-firmware/msg5853727/#msg5853727

Thank you.

[tautech]

[...] There is a small confirmed bug in last firmware (they are working on a fix) that if you are using DHCP, and Ethernet Cable is plugged in, if you cannot reach DHCP server on boot it can get stuck on Start screen[...]

Where is this reported?

Siglent private forum.

[2N3055]

[...] There is a small confirmed bug in last firmware (they are working on a fix) that if you are using DHCP, and Ethernet Cable is plugged in, if you cannot reach DHCP server on boot it can get stuck on Start screen[...]

Where is this reported?

Siglent private forum.

It was confirmed here and reported. To Siglent, who else? I don't think reporting it to Keysight would do us much good.

[KungFuJosh]

It was confirmed here and reported. To Siglent, who else? I don't think reporting it to Keysight would do us much good.

Probably wouldn't do us any good even if it was a Keysight product. 😉😉