NTP Gone MAD

Yes on my pro (just posted it on a different thread).

Though not my Duo and NV+, which used different NTP (the Pro used time-a.netgear.com, the others uses time-e.netgear.com, which apparently is ok).

Yes, same here. Transmission was hosed as a result and loads of my timestamps are now off.

They still haven't fixed the problem, and several hours have passed. Pretty lame.

I've now changed the NTP servers to use a more reliable source.

I had some script issues (all my backups were done in 1992 ;) ) and noticed the same thing.

Even if I switched to different netgear time servers, the time was off by an hour. I had noticed this previously but missed the daylight savings issue. I've also switched to the same time server(s) used in Windows instead.

[Edit] My back up script has been off my an hour for weeks. I don't have a specific date for when it started, but it was running an hour behind schedule. As what would seem to be a completely unrelated note, iTunes was starting to delete songs when I would sync them to my iPod. :shock: There didn't seem to be any rhyme or reason as to how, but various songs would get deleted during the sync process. After resetting my time server to a source that provides the correct time (no longer off by an hour, much less in 1992!) that issue disappeared. So there is apparently something in iTunes that was being thrown off by the one hour difference between my local computer time and the netgear server time.

Go figure.

Without the "1992" issue, I never would have figured this out.

How did you change which time servers you were using?
I can't find ntp.conf

EDIT: don't worry I found the right file

Just hit this as well, switched to http://www.pool.ntp.org/zone/north-america

I wish companies would just use the public ntp pool instead of running their own services. Though Netgear does have some neat history with NTP ;)

http://pages.cs.wisc.edu/~plonka/netgear-sntp/

I had the same problem. Switched over to http://www.pool.ntp.org/zone/au as others have done and all is working again.

For anyone interested in working out the time of failure, consecutive timestamps from something I was logging (with ~10-20 seconds between each line) were:

2012-04-08 13:34:10
1992-08-23 13:34:22

I think these are GMT+11 (or GMT+10 — I corrected my timezone in the RAIDiator web admin after this, so if it was somehow obeying daylight savings without my assistance it'd be +10). Though I'm not sure if the ReadyNAS uses ntpd or ntpdate, and not sure what the update frequency might have been...

Anyway, it seems to have just been a change in date, but not the time. According to Wolfram Alpha (link at bottom for the query I used) its almost exactly 1024 weeks between those timestamps... in fact, it's just 12 seconds short, which is likely the actual time between lines in the log ;)

It's a strange error... I'd have expected roll-over on 2^8 or 2^16 not 2^10... and something in seconds, days or years... not weeks. I wonder what the source of netgear's time is... and whether they're using ntpd or something custom...

Wolfram Alpha Query: http://www.wolframalpha.com/input/?i=1992-08-23+13%3A34%3A22++to+2012-04-08+13%3A34%3A10

This is a known issue: http://www.readynas.com/forum/viewtopic.php?p=349868#p349868

On Sparc and x86 you should be able to go to System > Clock in Frontview and change the NTP servers that are used.

Wish wrote:
How did you change which time servers you were using? I can't find ntp.conf EDIT: don't worry I found the right file

I changed my settings via the admin front-end, under the System tab there is an option to set the date and time, and also to enter a couple of time server addresses.

The Dashboard on the new ARM devices (Duo/NV+ v2) is different and it doesn't have an option to set the NTP servers. It just has the option to synchronisation with an NTP server chosen by the NAS or set the time manually.

At the moment the inability to configure the NTP servers seems like a bad idea... (This is also the case on most newer Netgear routers).