Samba/SMB not connecting. Telnet/unmenu/web fine

[benyaki]

I have been running this server for a month or so now with no problems.

I have just 6 drives in it now, running version 4.7.

Just yesterday the SMB shares started to be unreachable. I only have macs in the house, and connecting to them does not error out, it just keeps searching and searching.

I am running a few packages and scripts. Most recently added (less than a week ago) was a S3 sleep script (from the forum). Added line to go script.

Also running UPS package for APC UPS. Have had no problems with that.

Tried setting up airvideo a week or two ago to no avail, gave up and removed anything that was added to the USB stick.

I ran checkdisk on the USB stick today (through parallels desktop on a mac).

I have run memtest overnight (14+ hours) with 9 or 10 passes and no errors.

This morning I recopied bzimage and bzroot to the drive to make sure they were not corrupt and the problem persists.

The server is setup with a static IP from day 1 with no problems.

I was briefly able to connect to the SMB share this morning RIGHT after booting, but within a few seconds it was again not connecting on the same computer and another mac.

 

I would appreciate any help with this. I am quite new to unraid but have been reaching as much as possible and have not found a solution to this on the forums (what has worked for other so far has not worked for me).

 

If there is any required information that was not included, please let me know.

 

Motherboard: MSI 880GMA-E53

CPU: Sempron 140 (downclocked to 800Mhz)

PSU: Corsair 650W TX V2

RAM: Kingston ValueRAM (KVR1333D3N9/2G)

USB: Lexar firefly 4GB

 

Syslog Attached.

syslog.txt

[defected07]

Can you reboot your server, and copy the log after you've experienced connection issues over Samba? I only see two lines with 'smb' in them, and they both appear to be during start up/refresh of Samba.

 

Thanks

[benyaki]

That log was made after it started happening (made this morning after it could not connect).

I was initially able to access the share (but just did not mount it) on one of my macs. Then from another one of them, it just kept waiting for the connection from the server.

I will do again what you have asked and post it (will be a hour or so at the earliest).

 

Thanks

[defected07]

Yeah I just don't see anything, but then again, I have problems with AFP on my server, and the log isn't very good.

 

Take a look at your Console on your Mac, and see if that outputs anything about trying to mount your network drives, also.

[benyaki]

In console on both machine that have tried to access the server, there is nothing in console aside from one machine reporting unable to backup time machine (because the drive is currently powered down) and the other with some gfxcardstatus reports of switching to the GPU, nothing is out of place.

 

I am at work now and cannot seem to VNC back to my mini at home (was installing updates just before I left) so I will get another syslog this evening.

 

Thanks

[cyrnel]

Can your Parallels Windows use the shares?

 

(drive-by thinking)

[benyaki]

Tried with parallels but no go.

While thinking about what had changed within the last week, I focused in on the S3 script that I added. Didn't think much about it at first, but when I start up the server, as soon as SAMBA starts (hooked up to monitor, watching it load) I can access the share. Then as soon as packages are done installing and the sleep script starts counting, it does not return a login line or anything like that, the last line just says "COUNT 15" until the drives are spun up, then down, and then it repeats (the 15 minutes after idle it is set to go to sleep.

So I just removed the script, the line from the go script, and everything seems to be working OK now.

I really don't understand WHY ... but I guess unless someone has an idea, I am going to say this was the problem and I will have to do without sleep.

 

I will update this in a day or so, but I would say this is tentatively solved.

[defected07]

What's the sleep script, and what's the line you added to the go script?

[benyaki]

line added to go script directly after last line in go script

 

"/boot/custom/bin/s3.sh"

 

and the script in the appropriate folder (attached as a .txt, will not let me attach a sh file)

s3.txt

[benyaki]

If someone has a look at this and can let me know if anything is wrong with the S3 script or implementation, that would be great. I would like to save some power by having this sleep when not in use.