<div dir="ltr">Hi Daniel,<br><br>- increase MAXAGE in upsd.conf<br>- increase DEADTIME in upsmon.conf<br>- possibly increase the UPS polling interval (pollinterval in ups.conf) which is set to 2 seconds currently, to lower the communication load on the UPS.<br>
<br>the exact values depends on what you see in your log.<br>ie, if your UPS is marked stale for 20 seconds, putting 25 for MAXAGE and DEADTIME should be fine.<br><br>-- Arnaud<br><br><div class="gmail_quote">2008/7/22 Daniel O'Connor <<a href="mailto:doconnor@gsoft.com.au">doconnor@gsoft.com.au</a>>:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Hi,<br>
I have the above UPS connected to a FreeBSD 6.2 box and it shows 'data<br>
stale' very frequently (every few minutes).<br>
<br>
I have tried the usual trick of setting maxage, poll interval, etc etc<br>
but to no avail.<br>
<br>
When I wind up logging I see (on mge-shut)<br>
<br>
entering shut_get_report(id: 0d, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 16, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 17, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 2f, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 20, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 07, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 0a, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 07, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 04, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 12, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 04, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 12, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 13, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 39, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 37, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 38, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 04, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 12, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 13, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 3e, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 3c, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 3d, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 06, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering shut_get_report(id: 06, len: 1800)<br>
shut_wait_ack(): ACK received<br>
entering upsdrv_updateinfo()<br>
entering shut_get_report(id: 0f, len: 1800)<br>
<br>
And upsd..<br>
write: [destfd=7] [len=15] [ERR DATA-STALE]<br>
sstate_dead: driver for UPS [ups1] says data is stale<br>
<br>
Broadcast Message from <a href="mailto:radar@negrocreek.dnsalias.net">radar@negrocreek.dnsalias.net</a><br>
(no tty) at 7:57 UTC...<br>
<br>
Communications with UPS ups1@localhost lost<br>
<br>
sstate_dead: driver for UPS [ups1] says data is stale<br>
UPS [ups1] data is no longer stale<br>
acl_check: localhost: match 1<br>
ACL [localhost] matches, action=1<br>
write: [destfd=7] [len=30] [VAR ups1 ups.status "OL CHRG"]<br>
<br>
Broadcast Message from <a href="mailto:radar@negrocreek.dnsalias.net">radar@negrocreek.dnsalias.net</a><br>
(no tty) at 7:57 UTC...<br>
<br>
Communications with UPS ups1@localhost established<br>
<br>
acl_check: localhost: match 1<br>
ACL [localhost] matches, action=1<br>
<br>
It seems to decide very quickly that things are stuffed :(<br>
<br>
I just built a fresh NUT 2.2.2 (from 2.2.0) but got the same problem.<br>
<br>
Anyone have any suggestions?<br>
<br>
It does not seem related to the known issue in the mge-shut man page but<br>
rather that the UPS does not respond properly (or the OS is dropping<br>
data but I don't see any errors about it).<br>
<br>
One other possibility is that noise is being seen on the serial line<br>
(this system drives a high power VHF radar) - I have asked someone<br>
onsite to check the cabling but it would be nice if NUT could be<br>
adjusted to be more tolerant.<br>
<br>
Thanks.<br>
<font color="#888888"><br>
--<br>
Daniel O'Connor software and network engineer<br>
for Genesis Software - <a href="http://www.gsoft.com.au" target="_blank">http://www.gsoft.com.au</a><br>
"The nice thing about standards is that there<br>
are so many of them to choose from."<br>
-- Andrew Tanenbaum<br>
GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C<br>
</font><br>_______________________________________________<br>
Nut-upsuser mailing list<br>
<a href="mailto:Nut-upsuser@lists.alioth.debian.org">Nut-upsuser@lists.alioth.debian.org</a><br>
<a href="http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser" target="_blank">http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser</a><br></blockquote></div><br></div>