RPi3A+ Network instability. (2024)

zandman

RPi3A+ Network instability. (2)


Junior Member
RPi3A+ Network instability. (3)

Posts: 3
Threads: 1
Joined: Mar 2024

Reputation:

#11

06-14-2024, 11:50 AM (This post was last modified: 06-14-2024, 12:29 PM by zandman.)

I did have similar issues with my Pi3A+, but only on 9.0.1. You are not alone :-)
It starts up fine from a fresh SD image, but after changing some essential settings the interface freezes. I believe it happens after I add my NAS tot the library
No issues on 8.3.9

Today my Pi didn't even startup anymore, it doesn't read the SD card. So maybe my issues were (partially) due to a failing SD slot.

New Pi5 is on the way

Find

Reply

TheOldPresbyope


Moderator
RPi3A+ Network instability. (6)RPi3A+ Network instability. (7)RPi3A+ Network instability. (8)RPi3A+ Network instability. (9)

Posts: 5,699
Threads: 169
Joined: Apr 2018

Reputation:

216

#12

06-14-2024, 01:34 PM (This post was last modified: 06-14-2024, 01:46 PM by TheOldPresbyope. Edit Reason: ETA ping statistics)

@the_bertrum

With moOde 9.0.2 my Pi 3A+ in the sunroom is rock solid. BT instability has disappeared; loading an SMB share proceeds no matter how much futzing I do with the WebUI (such as rapidly switching among radio stations); yada yada yada.

The Pi is about 35 feet from the router/access point, more or less line-of-sight through two open doorways in a wood-framed house. One item which could make a difference: there's two dozen or so2.4 GHz WiFi access points in our neighborhood with many clients, so I'm using the 5 GHz band to minimize interference. Many sources there too but interference falls off faster with distance at 5 GHz

The SMB share (500 albums/6000 tracks) is offered by an Open Media Vault server running on a HardKernel HC1 ARM-based board connected to the router via Ethernet (with a segment via PowerLine Networking).

Stats from ifconfig and iwconfig on the Pi

Code:

ifconfig:
...
ether b8:27:eb:xx:xx:xx txqueuelen 1000 (Ethernet)
RX packets 865144 bytes 1109644913 (1.0 GiB)
RX errors 0 dropped 7 overruns 0 frame 0
TX packets 433087 bytes 79240364 (75.5 MiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

iwconfig:
...
Mode:Managed Frequency:5.785 GHz Access Point: A8:70:5D:xx:xx:xx
Bit Rate=390 Mb/s Tx-Power=31 dBm
Retry short limit:7 RTS thr:off Fragment thr:off
Power Management:off
Link Quality=67/70 Signal level=-43 dBm
Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
Tx excessive retries:107 Invalid misc:0 Missed beacon:0

ETA -

Forgot to include ping statistics.

Code:

--- 10.0.0.1 ping statistics ---
100 packets transmitted, 100 received, 0% packet loss, time 99151ms
rtt min/avg/max/mdev = 1.242/5.284/22.307/3.086 ms

I'm at a loss to explain your experience.

Regards,
Kent

Find

Reply

the_bertrum


Senior Member+
RPi3A+ Network instability. (12)

Posts: 1,638
Threads: 39
Joined: Mar 2020

Reputation:

77

(06-14-2024, 11:50 AM)zandman Wrote: I did have similar issues with my Pi3A+, but only on 9.0.1. You are not alone :-)
It starts up fine from a fresh SD image, but after changing some essential settings the interface freezes. I believe it happens after I add my NAS tot the library
No issues on 8.3.9

Today my Pi didn't even startup anymore, it doesn't read the SD card. So maybe my issues were (partially) due to a failing SD slot.

New Pi5 is on the way

What NAS are you using @zandman?

----------------
Robert

Find

Reply

the_bertrum


Senior Member+
RPi3A+ Network instability. (15)

Posts: 1,638
Threads: 39
Joined: Mar 2020

Reputation:

77

#14

06-18-2024, 02:34 PM

Ok, so here's my stats and some musings from several days of messing about.

A completely clean system with no configuration other than adding the SMB share. Doing nothing we get pings to the NAS (claudia.local) like this:

Code:

--- claudia.local ping statistics ---
100 packets transmitted, 100 received, 0% packet loss, time 99151ms
rtt min/avg/max/mdev = 2.331/5.264/10.978/1.334 ms

The network stats look like this:

Code:

aster@testbed:~ $ iwconfig
lo no wireless extensions.

wlan0 IEEE 802.11 ESSID:"TeawithRuby"
Mode:Managed Frequency:2.412 GHz Access Point: xx:xx:xx:xx:xx:xx
Bit Rate=72.2 Mb/s Tx-Power=31 dBm
Retry short limit:7 RTS thr:off Fragment thr:off
Power Management:off
Link Quality=49/70 Signal level=-61 dBm
Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
Tx excessive retries:0 Invalid misc:0 Missed beacon:0

master@testbed:~ $ ifconfig
lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
inet6 ::1 prefixlen 128 scopeid 0x10<host>
loop txqueuelen 1000 (Local Loopback)
RX packets 493 bytes 42733 (41.7 KiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 493 bytes 42733 (41.7 KiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

wlan0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 192.168.1.73 netmask 255.255.255.0 broadcast 192.168.1.255
inet6 fe80::e37c:6082:21f3:16fe prefixlen 64 scopeid 0x20<link>
ether b8:27:eb:39:c6:5a txqueuelen 1000 (Ethernet)
RX packets 10615 bytes 4576639 (4.3 MiB)
RX errors 0 dropped 4 overruns 0 frame 0
TX packets 3800 bytes 4264896 (4.0 MiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

So far so good, now we kick off a scan and ping looks like this:

Code:

--- claudia.local ping statistics ---
100 packets transmitted, 100 received, 0% packet loss, time 99462ms
rtt min/avg/max/mdev = 1.398/10.543/61.529/10.859 ms

Average doubled, but no packet loss or anything. I'd expect it to be slower, there's more going on on the network after all.
Network stats don't look any different really either. Tis from after the scan had run for a minute or so and I'd run 100 pings:

Code:

master@testbed:~ $ iwconfig
lo no wireless extensions.

wlan0 IEEE 802.11 ESSID:"TeawithRuby"
Mode:Managed Frequency:2.412 GHz Access Point: xx:xx:xx:xx:xx:xx
Bit Rate=65 Mb/s Tx-Power=31 dBm
Retry short limit:7 RTS thr:off Fragment thr:off
Power Management:off
Link Quality=50/70 Signal level=-60 dBm
Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
Tx excessive retries:0 Invalid misc:0 Missed beacon:0

master@testbed:~ $ ifconfig
lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
inet6 ::1 prefixlen 128 scopeid 0x10<host>
loop txqueuelen 1000 (Local Loopback)
RX packets 1320 bytes 103290 (100.8 KiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 1320 bytes 103290 (100.8 KiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

wlan0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 192.168.1.73 netmask 255.255.255.0 broadcast 192.168.1.255
inet6 fe80::e37c:6082:21f3:16fe prefixlen 64 scopeid 0x20<link>
ether b8:27:eb:39:c6:5a txqueuelen 1000 (Ethernet)
RX packets 292022 bytes 415853125 (396.5 MiB)
RX errors 0 dropped 6 overruns 0 frame 0
TX packets 108325 bytes 13774061 (13.1 MiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

I didn't have anything playing and wasn't messing in the UI at all during this run of the test. I did check the mpd log for activity while the scan was running every so often, and re-ran the network commands a few times. This last was seconds before the system froze:

Code:

master@testbed:~ $ ifconfig
lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
inet6 ::1 prefixlen 128 scopeid 0x10<host>
loop txqueuelen 1000 (Local Loopback)
RX packets 2810 bytes 211474 (206.5 KiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 2810 bytes 211474 (206.5 KiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

wlan0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 192.168.1.73 netmask 255.255.255.0 broadcast 192.168.1.255
inet6 fe80::e37c:6082:21f3:16fe prefixlen 64 scopeid 0x20<link>
ether b8:27:eb:39:c6:5a txqueuelen 1000 (Ethernet)
RX packets 1076258 bytes 1574774233 (1.4 GiB)
RX errors 0 dropped 7 overruns 0 frame 0
TX packets 373034 bytes 36784306 (35.0 MiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

Again, nothing shouting out at me. The irritating thing is that even if there was a smoking gun in these stats when the issue hits, I can't see it because of the frozen state.

Other things I have tried - thinking it might be mDNS related somehow, I mounted the NAS using the IP reserved for it in the router's DHCP. Same behaviour. The share never shows up in the "Scan" in the moode configuration screen (never has on any version), so something isn't 100% in the exposure of the share on the network, but it does always mount when the path is added manually, and it works happily on my Pi4 9.x devices (one ethernet, one wifi) and worked fine on all 8.x and earlier versions (to about 6 when I started). I've used some tools on my Android phone to check the mDNS broadcast and _smb._tcp scans find the NAS correctly. Maybe there's mileage in my trying the moode scan code manually in case that shed some light? But then, mDNS is probably not the problem since it freezes with the IP as well. It's straws I'm clutching at.

One thing I can be pretty certain of is that it is related to the SMB share on the NAS. If I don't mount, scan or play from the NAS the system is fine. The swapping about in the UI or swapping stations was co-incidental, just a scan will kick off the issue most times. Playing from the NAS on the rare occasion that the scan doesn't freeze will eventually trigger it. I've listened to the radio for hours with no issue. I suspect it must be something the Synology does at it's end that is the trigger, but I'm at a loss to work out what that might be. I've turned off the disc power management just in case so the discs never sleep. Long shot since why would the discs sleep while a scan was underway? Anyway, it achieved nothing.

Next payday, I'll get a zero 2 and see if that plays ball and maybe try again everytime there's a kernel update in case. I can't see anyway to get any useful diagnostics for anyone to look at so I'll thank you for your time and move on I think. The Pi3 will have to find another job RPi3A+ Network instability. (16)

----------------
Robert

Find

Reply

Tim Curtis

RPi3A+ Network instability. (18)


moOde audio founder
RPi3A+ Network instability. (19)RPi3A+ Network instability. (20)RPi3A+ Network instability. (21)RPi3A+ Network instability. (22)RPi3A+ Network instability. (23)RPi3A+ Network instability. (24)

Posts: 12,453
Threads: 278
Joined: Mar 2018

Reputation:

487

#15

06-18-2024, 05:48 PM

Couple other things to try:

1. Monitor the system journal and see if the last message before the systemhangs is useful

Code:

journalctl -f

2. Configure your NAS for NFS instead of Samba and see ifthat fixes the issue. Thats what I use in my config.

Enjoy the Music!
moodeaudio.org | Mastodon Feed | GitHub

Website

Find

Reply

the_bertrum


Senior Member+
RPi3A+ Network instability. (27)

Posts: 1,638
Threads: 39
Joined: Mar 2020

Reputation:

77

#16

Yesterday, 07:06 AM

(06-18-2024, 05:48 PM)TimCurtis Wrote: Couple other things to try:

1. Monitor the system journal and see if the last message before the systemhangs is useful

Code:

journalctl -f


2. Configure your NAS for NFS instead of Samba and see ifthat fixes the issue. Thats what I use in my config.

I've been tailing the journal like a man possessed since this issue arose. The last entry isn't consistent, sometimes it's a "timedate" job, sometimes a system clean up job, most often the freeze is in the pauses between things getting written.

I'll try NFS, I always felt NFS would be the sensible choice between two Linux systems, but back when I first got moode, about version 6, I found it would unmount regularly. Of course we have a mount monitor now RPi3A+ Network instability. (28)

----------------
Robert

Find

Reply

the_bertrum


Senior Member+
RPi3A+ Network instability. (31)

Posts: 1,638
Threads: 39
Joined: Mar 2020

Reputation:

77

#17

Yesterday, 08:15 PM

Right, so more investigation needed to conform a few things but....

I tired NFS and was sad to see it got stuck again, so I reset it and tried again while tailing the journal in case NFS showed something up that SMB didn't. It only went and ran to completion! So I ran a rebuild of the library and it completed again. I noticed some errors relating to not having permissions in the "@eaDIR" directories, which are annoying folders that DSM uses for it's own purposes. So I put an mpdignore file in the root with the directory name in it to exclude them and ran another update, the errors disappeared as expected, so I swapped back to SMB just for information. It worked then too! I need a few more tests to be sure, but it may just be those directories causing something to crash.

----------------
Robert

Find

Reply

Tim Curtis

RPi3A+ Network instability. (33)


moOde audio founder
RPi3A+ Network instability. (34)RPi3A+ Network instability. (35)RPi3A+ Network instability. (36)RPi3A+ Network instability. (37)RPi3A+ Network instability. (38)RPi3A+ Network instability. (39)

Posts: 12,453
Threads: 278
Joined: Mar 2018

Reputation:

487

#18

Yesterday, 08:56 PM

Those hidden extended directory attribute files and dirs are really annoying. Windows has a bunch of them, MacOS has the their super annoying resource fork hidden files. I don't know of Linux desktop apps (KDE, Gnome) also create these hidden files.

Can you post an ls -al or similar so I can see what these files look like?
Maybe I can try creating some on my Pi file server and see what happens.

Enjoy the Music!
moodeaudio.org | Mastodon Feed | GitHub

Website

Find

Reply

the_bertrum


Senior Member+
RPi3A+ Network instability. (42)

Posts: 1,638
Threads: 39
Joined: Mar 2020

Reputation:

77

#19

7 hours ago (This post was last modified: 6 hours ago by the_bertrum.)

Code:

root@Claudia:/volume1/music/The Replacements/All Shook Down# ls -la
total 39884
drwxrwxrwx+ 3 CloudAdmin users 4096 Feb 2 2020 .
drwxrwxrwx+ 3 CloudAdmin users 4096 Apr 25 2018 ..
-rwxrwxrwx+ 1 CloudAdmin users 3635629 Feb 2 2020 '01 Merry Go Round.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 2962444 Feb 2 2020 '02 One Wink At A Time.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 3151908 Feb 2 2020 '03 Nobody.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 3596670 Feb 2 2020 '04 Bent Out Of Shape.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 3137552 Feb 2 2020 '05 Sadly Beautiful.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 3671712 Feb 2 2020 '06 Someone Take The Wheel.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 3112614 Feb 2 2020 '07 When It Began.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 3201763 Feb 2 2020 '08 All Shook Down.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 2671864 Feb 2 2020 '09 Attitude.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 2877728 Feb 2 2020 '10 Happy Town.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 1893491 Feb 2 2020 '11 Torture.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 3950017 Feb 2 2020 '12 My Little Problem.m4a'
-rwxrwxrwx+ 1 CloudAdmin users 2856709 Feb 2 2020 '13 The Last.m4a'
drw-rw-r-- 17 root root 4096 Aug 29 2020 @eaDir
-rwxrwxrwx+ 1 root root 53532 Feb 2 2020 folder.jpg
-rwxrwxrwx+ 1 CloudAdmin users 18922 Apr 5 2017 Folder.jpg
root@Claudia:/volume1/music/The Replacements/All Shook Down# cd \@eaDir/
root@Claudia:/volume1/music/The Replacements/All Shook Down/@eaDir# ls -la
total 68
drw-rw-r-- 17 root root 4096 Aug 29 2020 .
drwxrwxrwx+ 3 CloudAdmin users 4096 Feb 2 2020 ..
drw-rw-r-- 2 root root 4096 Sep 2 2021 '01 Merry Go Round.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '02 One Wink At A Time.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '03 Nobody.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '04 Bent Out Of Shape.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '05 Sadly Beautiful.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '06 Someone Take The Wheel.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '07 When It Began.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '08 All Shook Down.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '09 Attitude.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '10 Happy Town.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '11 Torture.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '12 My Little Problem.m4a'
drw-rw-r-- 2 root root 4096 Sep 2 2021 '13 The Last.m4a'
drw-rw-r-- 2 root root 4096 Dec 15 2019 folder.jpg
drw-rw-r-- 2 root root 4096 Aug 29 2020 Folder.jpg
root@Claudia:/volume1/music/The Replacements/All Shook Down/@eaDir# cd 01\ Merry\ Go\ Round.m4a/
root@Claudia:/volume1/music/The Replacements/All Shook Down/@eaDir/01 Merry Go Round.m4a# ls -la
total 124
drw-rw-r-- 2 root root 4096 Sep 2 2021 .
drw-rw-r-- 17 root root 4096 Aug 29 2020 ..
-rwxrwxrwx 1 CloudAdmin users 53532 Feb 1 2021 01APIC_03.jpg
-rw-r--r-- 1 CloudAdmin users 53532 Sep 2 2021 SYNOAUDIO_01APIC_03.jpg

I want to run some more rigorous tests before I conclude these directories are to blame. I need to be a bit more systematic in running the update from a clean system with and without the mpdignore mask.
Eitherway, that was listing from on the NAS itself (what's with that odd mix of owners?), if a listing from the /mnt on the pi would be of use too, I'll get one next time I generate with the mpdignore removed.

ETA: of course I know mpdignore has ne effect on the mount. Dunno why I wrote that. Not enough coffee.

----------------
Robert

Find

Reply

Tim Curtis

RPi3A+ Network instability. (44)


moOde audio founder
RPi3A+ Network instability. (45)RPi3A+ Network instability. (46)RPi3A+ Network instability. (47)RPi3A+ Network instability. (48)RPi3A+ Network instability. (49)RPi3A+ Network instability. (50)

Posts: 12,453
Threads: 278
Joined: Mar 2018

Reputation:

487

#20

4 hours ago

Is there such a thing as too much coffee?

Enjoy the Music!
moodeaudio.org | Mastodon Feed | GitHub

Website

Find

Reply

RPi3A+ Network instability. (2024)
Top Articles
Latest Posts
Article information

Author: Twana Towne Ret

Last Updated:

Views: 6313

Rating: 4.3 / 5 (64 voted)

Reviews: 95% of readers found this page helpful

Author information

Name: Twana Towne Ret

Birthday: 1994-03-19

Address: Apt. 990 97439 Corwin Motorway, Port Eliseoburgh, NM 99144-2618

Phone: +5958753152963

Job: National Specialist

Hobby: Kayaking, Photography, Skydiving, Embroidery, Leather crafting, Orienteering, Cooking

Introduction: My name is Twana Towne Ret, I am a famous, talented, joyous, perfect, powerful, inquisitive, lovely person who loves writing and wants to share my knowledge and understanding with you.