Firmware 4.05.00-315 Discussion

Ah yep missed that. Much better sleep time. :smiley:

Same problem with wake-ups here. An example of my "sleep"times:

04 29 17:32:30 17:37:19   289  0:04:49
04 29 17:47:30 17:52:20   289  0:04:49
04 29 18:02:30 18:07:20   289  0:04:49
04 29 18:17:30 18:22:20   290  0:04:50
04 29 18:32:30 18:37:19   289  0:04:49
04 29 18:47:30 18:52:20   290  0:04:50
04 29 19:02:30 19:07:20   290  0:04:50
04 29 19:17:30 19:22:20   290  0:04:50
04 29 19:32:30 19:37:19   289  0:04:49
04 29 19:47:30 19:52:20   289  0:04:49
04 29 20:02:30 20:07:19   289  0:04:49
04 29 20:17:30 20:22:19   289  0:04:49
04 29 20:32:30 20:37:19   289  0:04:49
04 29 20:47:30 20:52:20   290  0:04:50
04 29 21:02:30 21:07:19   289  0:04:49
04 29 21:17:30 21:22:20   290  0:04:50
04 29 21:32:30 21:37:20   290  0:04:50
04 29 21:47:30 21:52:20   290  0:04:50
04 29 22:02:30 22:07:20   289  0:04:49
04 29 22:17:30 22:22:20   289  0:04:49
04 29 22:32:30 22:37:20   289  0:04:49
04 29 22:47:30 22:52:19   289  0:04:49
04 29 23:02:30 23:07:20   290  0:04:50
04 29 23:17:30 23:22:19   289  0:04:49
04 29 23:32:30 23:37:19   289  0:04:49
04 29 23:47:30 23:52:19   289  0:04:49
04 30 00:02:30 00:07:20   290  0:04:50
04 30 00:17:30 00:22:20   290  0:04:50
04 30 00:32:30 00:37:19   289  0:04:49
04 30 01:42:25 01:42:33     7  0:00:07

Does anyone know what causes these wake-ups? I tried searching all log files, but couldn’t find anything.
And if it’s samba, should i downgrade the firmware?

Did some quick and dirty testing on my end with a first gen My Cloud running 4.05.00-315 to investigate the lack of sleep issue. I can basically/generally confirm Rac8006 findings that disabling Samba does indeed cause the device to sleep dramatically more than when Samba is running.

With no modifications in the /etc/rc2.d/S98user-start file, with remote access disabled in the Dashboard, with Samba running. And no devices accessing the My Cloud.

/DataVolume/shares/xxx/WD My Cloud$ sh sleeptime.sh
04 30 09:34:07 09:34:15     8  0:00:08
04 30 09:44:27 09:44:39    12  0:00:12
04 30 09:54:51 09:56:29    98  0:01:38
04 30 10:22:00 10:22:07     7  0:00:07
04 30 10:32:19 10:41:30   551  0:09:11
04 30 10:51:42 10:56:30   288  0:04:48
Total Sleep Time:  0:00:00
Start 1493553726
End 1493565468
Total Up Time:  3:15:42  8% Sleep 92% Wake

Now, the same but with Samba disabled (/etc/rc2.d/S20samba stop) and no devices accessing the My Cloud.

/DataVolume/shares/xxx/WD My Cloud$ sh sleeptime.sh
04 30 12:06:40 12:41:34  2094  0:34:54
04 30 12:51:46 15:17:10  8724  2:25:24
04 30 15:27:22 16:01:09  2027  0:33:47
04 30 16:11:22 17:07:23  3361  0:56:01
Total Sleep Time:  4:00:00
Start 1493568128
End 1493586493
Total Up Time:  5:06:05 88% Sleep 12% Wake

With Samba disabled the wake/sleep values are practically reversed. With Samba enabled, unit slept only 16 minutes in the space of 3 hours and 15 minutes (approx). Compared with Samba stopped the unit slept for 4 hours and 30 minutes in the space of 5 hours (approx).

This begs the question is there something with the change to SMB3 that was instituted in this firmware release that is causing the lack of sleep or is it some other issue with Samba/firmware settings? Obviously one possible way to test,is edit the /etc/samba/smb-global file and change max protocol = SMB3 to max protocol = SMB2 and see if that has any effect.

Proceeded to install this firmware today on my WD My Cloud and it pooched during the process.
It just timed out close to 10 mins of trying to install the update. First time I’ve ever had such an issue.
It seems that it does eventually boot, but when I try to access the UI via web, I get a quick error that the username/password is invalid, but not sure what that error is about and therefore can’t login.

Anyone know how I can get this thing up and running again?

Changing to smb2 and restart didn’t help. I stopped all cron jobs an will see tomorrow…

Wait for @Bill_S to contact you and they will trade your hard drive full of data for a refurbished empty one.

or… from our new first level support from WD…

After some more testing I’ve found that after rebooting my other My Cloud. Which has not been rebooted since I upgraded the firmware on the failing My Cloud. The one minute entries in the /var/log/samba/log.smbd stopped.
I need to leave the system alone for a few hours to see if the sleep problem is corrected.

you know rac8006, I have to say… I wonder about you sometimes… :stuck_out_tongue:

You did one restart - during the FW upgrade :wink:

If you disable Samba is it still possible to access My Cloud using either Windows explorer or MacOS Finder?

Thanks for the info Ralphael. I was actually able to recover using the 40 sec reset trick. After that I manually flashed again for good measure and things appear to be running normal again with all my data intact.

Same here with SMB2. Let the My Cloud run for a couple of hours at max protocol = SMB2, still poor sleeping when My Cloud isn’t being used.

/DataVolume/shares/xxx/WD My Cloud$ sh sleeptime.sh
04 30 18:14:36 18:14:44     8  0:00:08
04 30 18:24:56 18:25:16    20  0:00:20
04 30 18:35:29 18:47:13   703  0:11:43
04 30 18:57:25 19:02:20   295  0:04:55
04 30 19:12:32 19:17:20   288  0:04:48
04 30 19:27:32 19:32:20   288  0:04:48
04 30 20:17:59 20:18:06     7  0:00:07
04 30 20:28:18 20:33:37   319  0:05:19
04 30 20:43:49 20:48:30   281  0:04:41
04 30 20:58:42 21:02:19   217  0:03:37
04 30 21:12:31 21:18:37   366  0:06:06
04 30 21:28:49 21:33:37   288  0:04:48
04 30 21:43:49 21:48:37   288  0:04:48
04 30 21:58:49 22:03:37   288  0:04:48
Total Sleep Time:  1:00:00
Start 1493588802
End 1493604336
Total Up Time:  4:18:54 23% Sleep 77% Wake

Why?

For Mac I assume (not have a Mac to test with) you could still access the My Cloud using AFS if one disabled Samba.

For Windows, if one has the right windows version that offers NFS, they might be able to use NFS to access the My Cloud if one disables Samba.

Spoken about bad sleeping-times: I’m still on v04.04.05-101 (with remote access ON and no processes stopped manually):

Total Sleep Time: 16:00:40
Start 1489456952
End 1493754929
Total Up Time: 1193:52:57 1% Sleep 99% Wake

The good news is … the latest firmware will not harm ‘my’ sleep-times to much :sweat:

This is my latest sleep time. Ignore the total line. I made a change around midnight. Notice the 4.48 times
are gone. What ever samba is doing is still happening but it does not effect the sleep time. What I did was
what we talked about in the past for fixing sleep time. I just forgot to make the change.

mount -o remount,noatime,nodiratime /dev/root /

05 01 22:25:31 22:30:12 281 0:04:41
05 01 22:40:24 22:45:12 288 0:04:48
05 01 22:55:24 23:00:12 288 0:04:48
05 01 23:10:24 23:15:12 288 0:04:48
05 01 23:25:24 23:26:07 43 0:00:43
05 02 00:05:58 00:06:06 8 0:00:08
05 02 00:16:19 00:17:06 47 0:00:47
05 02 00:27:18 00:30:06 168 0:02:48
05 02 00:42:27 00:42:35 8 0:00:08
05 02 00:52:52 01:01:17 505 0:08:25
05 02 01:11:35 01:18:45 430 0:07:10
05 02 01:29:03 01:47:44 1121 0:18:41
05 02 01:58:01 02:58:06 3605 1:00:05
05 02 03:15:26 03:15:34 8 0:00:08
05 02 03:25:51 03:50:47 1496 0:24:56
05 02 04:01:06 04:02:03 57 0:00:57
05 02 04:12:20 06:11:04 7124 1:58:44
05 02 06:21:21 06:29:00 459 0:07:39
05 02 06:39:18 06:56:37 1039 0:17:19
05 02 07:06:55 07:17:32 637 0:10:37
05 02 07:27:49 07:28:54 65 0:01:05
05 02 07:39:11 07:39:19 8 0:00:08
05 02 08:04:01 08:04:09 8 0:00:08
05 02 08:14:26 08:17:06 160 0:02:40
05 02 08:27:23 09:01:21 2038 0:33:58
05 02 09:11:39 11:20:35 7736 2:08:56
05 02 11:30:53 11:31:00 7 0:00:07
05 02 11:41:18 15:00:40 11961 3:19:21
05 02 15:10:57 16:31:48 4851 1:20:51
Total Sleep Time: 46:20:51
Start 1493280131
End 1493757108
Total Up Time: 132:29:37 34% Sleep 66% Wake

Nice longer sleep periods.

Still investing my sleep issues. So far the only way I’ve found to get my first gen to sleep more than 10 minutes is to turn off Samba within the unit. Otherwise it sleeps less than 10% to 15% of the time even though its not in use for hours. Reflashed the 315 firmware, did a 40 second reset, enabled/disabled remote access and a few other things earlier to no effect. Unit still just doesn’t seem to want to sleep for longer periods of time for some reason.

The problem is that samba has the propensity to cause files to update there modify time when nothing has changed.
That is what the mount command does. It stops the system from updating the atime and diratime. If you just type mount you can see that the sda4 partition has these options set.

I am running the mount string in the /etc/rc2.d/S98user-start file. Here is the full contents of that file:

mount -o remount,noatime,nodiratime /dev/root /
/etc/init.d/wdmcserverd stop
/etc/init.d/wdphotodbmergerd stop
/etc/rc2.d/S20restsdk-serverd stop

Still have sleep issues due to Samba. Are you running any other code lines in that startup file or are you running the mount command at a later time through some other method (like manually)?

I’m doing the mount manually. What is the output from the mount command. Just type mount

his is my output
/dev/root on / type ext3 (rw,noatime,nodiratime,errors=continue,user_xattr,acl,barrier=1,data=ordered)
sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
tmpfs on /run type tmpfs (rw,nosuid,noexec,relatime,size=40960k,mode=755)
tmpfs on /run/lock type tmpfs (rw,nosuid,nodev,noexec,relatime,size=40960k)
proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
tmpfs on /dev type tmpfs (rw,relatime,size=10240k,mode=755)
tmpfs on /run/shm type tmpfs (rw,nosuid,nodev,noexec,relatime,size=5120k)
devpts on /dev/pts type devpts (rw,nosuid,noexec,relatime,gid=5,mode=620)
fusectl on /sys/fs/fuse/connections type fusectl (rw,relatime)
tmpfs on /tmp type tmpfs (rw,relatime,size=102400k,nr_inodes=20480)
/dev/root on /var/log.hdd type ext3 (rw,relatime,errors=continue,user_xattr,acl,barrier=1,data=ordered)
ramlog-tmpfs on /var/log type tmpfs (rw,relatime,size=40960k)
/dev/sda4 on /DataVolume type ext4 (rw,noatime,nodiratime,user_xattr,barrier=0,data=writeback)
/dev/sda4 on /CacheVolume type ext4 (rw,noatime,nodiratime,user_xattr,barrier=0,data=writeback)
/dev/sda4 on /nfs/SmartWare type ext4 (rw,noatime,nodiratime,user_xattr,barrier=0,data=writeback)
/dev/sda4 on /nfs/Public type ext4 (rw,noatime,nodiratime,user_xattr,barrier=0,data=writeback)
/dev/sda4 on /nfs/Ronald type ext4 (rw,noatime,nodiratime,user_xattr,barrier=0,data=writeback)
/dev/sda4 on /nfs/TimeMachineBackup type ext4 (rw,noatime,nodiratime,user_xattr,barrier=0,data=writeback)
/dev/sda4 on /nfs/George type ext4 (rw,noatime,nodiratime,user_xattr,barrier=0,data=writeback)
nfsd on /proc/fs/nfsd type nfsd (rw,relatime)
/dev/sdb1 on /var/media/USB type vfat (rw,relatime,uid=65534,gid=1000,fmask=0000,dmask=0000,allow_utime=0022,codepage=cp437,iocharset=iso8859-1,shortname=mixed,utf8,errors=continue)
/dev/sdb1 on /nfs/USB type vfat (rw,relatime,uid=65534,gid=1000,fmask=0000,dmask=0000,allow_utime=0022,codepage=cp437,iocharset=iso8859-1,shortname=mixed,utf8,errors=continue)
nodev on /sys/kernel/debug type debugfs (rw,relatime)

No success here either:

Tried mount -o remount,noatime,nodiratime /dev/root / and still had the 287s wakeups…

05 03 06:39:40 06:44:27   287  0:04:47
05 03 06:54:40 06:59:27   287  0:04:47
05 03 07:11:42 07:14:27   165  0:02:45
05 03 07:24:40 07:29:27   287  0:04:47
05 03 07:39:40 07:44:27   287  0:04:47
05 03 07:54:40 07:59:27   287  0:04:47

@rac8006 can you please upload your monitrorio.sh? I modified it and still have even the 7 (or now 17s) wakeups…

            touch /tmp/standby
            enterStandbyTime=`date +%s`
            sync
            sync
            sleep 10
            echo "Enter standby"

Besides the list in Bennors script did you modify something else?