ReadyNAS 628x Interface hangs while change NFS/SMB permissions

IPv6 has never been enabled.  I have two interfaces with link, I enabled the second interface with an IP but didn't get a different result.  I also enabled Plex and didn't get a different result.  I had an iSCSI lun but since removed prior to my previous post.   I can make changes anywhere in the system without issue.  This is very specific to making changes to a shares.   

Is the OS reload destructive?  I'd prefer to avoid having to buy backup disks - the reinstall would actually cost $$$.

Oct 29 07:32:12 NAS01 systemd[1]: Reloading.
Oct 29 07:32:12 NAS01 systemd-sysv-generator[6450]: Overwriting existing symlink /run/systemd/generator.late/umountiscsi.service with real service.
Oct 29 07:32:12 NAS01 systemd[1]: [/lib/systemd/system/fvapp-plexmediaserver.service:10] Invalid escape sequences in line, correcting: "/bin/sh -c ' PLEX_MEDIA_SERVER_INFO_VENDOR=Netgear  PLEX_MEDIA_SERVER_INFO_DEVICE="$(rn_nml -g systeminfo | grep "<Model>" | awk -F "[<>]" "{print \$3}")"  PLEX_MEDIA_SERVER_INFO_MODEL="$(uname -m)"  PLEX_MEDIA_SERVER_INFO_PLATFORM_VERSION="ReadyNAS OS $(rn_nml -g systeminfo | grep "<Firmware_Version>" | awk -F "[<>]" "{print \$3}")"  /apps/plexmediaserver/Binaries/Plex\ Media\ Server'"
Oct 29 07:32:12 NAS01 apachectl[6457]: [Thu Oct 29 07:32:12.832160 2020] [so:warn] [pid 6460] AH01574: module cgi_module is already loaded, skipping
Oct 29 07:32:12 NAS01 apachectl[6457]: AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 192.168.1.10. Set the 'ServerName' directive globally to suppress this message
Oct 29 07:32:12 NAS01 systemd[1]: Stopped The Apache HTTP Server.
Oct 29 07:32:12 NAS01 systemd[1]: Starting The Apache HTTP Server...
Oct 29 07:32:12 NAS01 systemd[1]: Reloading.
Oct 29 07:32:12 NAS01 systemd-sysv-generator[6559]: Overwriting existing symlink /run/systemd/generator.late/umountiscsi.service with real service.
Oct 29 07:32:12 NAS01 systemd[1]: [/lib/systemd/system/fvapp-plexmediaserver.service:10] Invalid escape sequences in line, correcting: "/bin/sh -c ' PLEX_MEDIA_SERVER_INFO_VENDOR=Netgear  PLEX_MEDIA_SERVER_INFO_DEVICE="$(rn_nml -g systeminfo | grep "<Model>" | awk -F "[<>]" "{print \$3}")"  PLEX_MEDIA_SERVER_INFO_MODEL="$(uname -m)"  PLEX_MEDIA_SERVER_INFO_PLATFORM_VERSION="ReadyNAS OS $(rn_nml -g systeminfo | grep "<Firmware_Version>" | awk -F "[<>]" "{print \$3}")"  /apps/plexmediaserver/Binaries/Plex\ Media\ Server'"
Oct 29 07:32:13 NAS01 apachectl[6567]: [Thu Oct 29 07:32:13.222905 2020] [so:warn] [pid 6571] AH01574: module cgi_module is already loaded, skipping
Oct 29 07:32:13 NAS01 apachectl[6567]: AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 192.168.1.10. Set the 'ServerName' directive globally to suppress this message
Oct 29 07:32:13 NAS01 apache2[6571]: [ssl:warn] [pid 6571] AH01909: 192.168.1.10:443:0 server certificate does NOT include an ID which matches the server name
Oct 29 07:32:13 NAS01 systemd[1]: Started The Apache HTTP Server.
Oct 29 07:32:13 NAS01 apache2[6574]: [ssl:warn] [pid 6574] AH01909: 192.168.1.10:443:0 server certificate does NOT include an ID which matches the server name
Oct 29 07:32:21 NAS01 apache_access[6575]: Admin session started from 10.0.22.214

---

bradthur wrote:

IPv6 has never been enabled. 

 

---

It is enabled by default, and your logs were showing an IPv6 address.  The snippet in your most recent post doesn't show it though.

---

bradthur wrote:

This is very specific to making changes to a shares.   

I understand those are your symptoms.  But the bottom line is that apache normally isn't restarted when you make a change to share settings (and even if it is restarted, it shouldn't end up with the loop are seeing).

---

bradthur wrote:

 

 I'd prefer to avoid having to buy backup disks - the reinstall would actually cost $$$.

I disagree here. Data Recovery is what actually costs $$$, and often fails.  Backups do cost $, but if you care about your data, you need to make that investment.  RAID isn't enough to protect it. 

---

bradthur wrote:

 

Is the OS reload destructive?

Normally an OS reinstall is not destructive.  The process is describe on pages 146-147 here: https://www.downloads.netgear.com/files/GDC/READYNAS-100/ReadyNAS_%20OS6_Desktop_HM_EN.pdf

Be careful not to do the factory default (which is destructive).

The OS reinstall does a partial reinstall of the OS partition.  It also resets a couple of configuration settings:

1. The NAS admin password is reset to password
2. The network settings will be set back to default DHCP (and ipv6 will be re-enabled).
3. Volume quota is turned off (you can turn it back on via the volume settings wheel).

Note that even if you con't use quotas, it is useful to have volume quota turned on.  You get better information on the storage used in each share (and the amount used in snapshots of each share).

StephenB wrote:

---

bradthur wrote:

IPv6 has never been enabled. 

 

---

It is enabled by default, and your logs were showing an IPv6 address.  The snippet in your most recent post doesn't show it though.

 

Fair enough, from day one (June something) default setting for IPv6 was set to disabled.  Can't explain why it doesn't show it now.  Even when I mod a share. 

 

---

bradthur wrote:

This is very specific to making changes to a shares.   

I understand those are your symptoms.  But the bottom line is that apache normally isn't restarted when you make a change to share settings (and even if it is restarted, it shouldn't end up with the loop are seeing).

 

Shoulder shrug

 

 

---

bradthur wrote:

 

 I'd prefer to avoid having to buy backup disks - the reinstall would actually cost $$$.

I disagree here. Data Recovery is what actually costs $$$, and often fails.  Backups do cost $, but if you care about your data, you need to make that investment.  RAID isn't enough to protect it. 

 

Do you think you really undstand the big picture enough where you can perscribe your backup theology and make a determination I don't care about my data?  We need to keep this to the issue at hand which is, I bought a product and it's not working.  It's not my fault it doesn't work and we wouldn't be having this conversation if this product was working correctly. I've only had this online since June and there isn't anything complex about my configuration or usage... CIFS/NFS/Plex.  So maybe I should be asking for a refund.   The bottom line is yes at the risk of having to reseed 2 - 3 weeks of data, I would have to purchase additional disks.  

 

---

bradthur wrote:

 

Is the OS reload destructive?

Normally an OS reinstall is not destructive.  The process is describe on pages 146-147 here: https://www.downloads.netgear.com/files/GDC/READYNAS-100/ReadyNAS_%20OS6_Desktop_HM_EN.pdf

 

Assuming we are committing to the re-install... Other than me accidentally selecting factory default, what is the real risk of performing this opereration non-destructively?

 

 

Be careful not to do the factory default (which is destructive).

 

The OS reinstall does a partial reinstall of the OS partition.  It also resets a couple of configuration settings:

1. The NAS admin password is reset to password
2. The network settings will be set back to default DHCP (and ipv6 will be re-enabled).
3. Volume quota is turned off (you can turn it back on via the volume settings wheel).

Note that even if you con't use quotas, it is useful to have volume quota turned on.  You get better information on the storage used in each share (and the amount used in snapshots of each share).

---

---

bradthur wrote:
... It's not my fault it doesn't work ...

---

It's not my fault either.  Note I don't work for Netgear (and never have).  So take a breath.

The OS reinstall is intended to be safe.  But there is clearly something broken with your system, and I think there is some risk that whatever is broken could lead to data loss. 

I took the gamble and reloaded the OS.  The reload was pretty.   Here's the journalctl -f output.

Oct 29 18:59:59 NAS01 readynasd[5337]: Snapper SetConfig successfully.
Oct 29 18:59:59 NAS01 systemd[1]: systemd-journald-audit.socket: Cannot add dependency job, ignoring: Unit systemd-journald-audit.socket is masked.
Oct 29 18:59:59 NAS01 readynasd[5337]: Protocol 'cifs' is reloaded
Oct 29 18:59:59 NAS01 systemd[1]: Reloading Samba NMB Daemon.
Oct 29 18:59:59 NAS01 systemd[1]: Reloaded Samba NMB Daemon.
Oct 29 18:59:59 NAS01 systemd[1]: Reloading Samba SMB Daemon.
Oct 29 18:59:59 NAS01 systemd[1]: systemd-journald-audit.socket: Cannot add dependency job, ignoring: Unit systemd-journald-audit.socket is masked.
Oct 29 18:59:59 NAS01 systemd[1]: Reloaded Samba SMB Daemon.
Oct 29 18:59:59 NAS01 systemd[1]: Reloading Samba NMB Daemon.
Oct 29 18:59:59 NAS01 systemd[1]: Reloaded Samba NMB Daemon.

Much better.  The interface is much faster as well.

My thoughts on RCA is the 626 Out of Memory crash probably corrupted the os.  

---

bradthur wrote:

I took the gamble and reloaded the OS. Much better.  The interface is much faster as well.

---

Great!  

---

bradthur wrote:

I took the gamble and reloaded the OS.  Much better.  The interface is much faster as well.

 

---

Great!.  The journalctl output looks normal to me.