bbb
-
Posts
20 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by bbb
-
-
Jan 7 08:33:12 hqv pulseway: Error while registering computer:SOAP 1.1 fault SOAP-ENV:Server[no subcode]#012"Timeout"#012Detail: connect failed in tcp_connect() Jan 7 08:33:43 hqv pulseway: Error while registering computer:SOAP 1.1 fault SOAP-ENV:Server[no subcode]#012"Timeout"#012Detail: connect failed in tcp_connect() Jan 7 08:34:14 hqv pulseway: Error while registering computer:SOAP 1.1 fault SOAP-ENV:Server[no subcode]#012"Timeout"#012Detail: connect failed in tcp_connect() Jan 7 08:34:45 hqv pulseway: Error while registering computer:SOAP 1.1 fault SOAP-ENV:Server[no subcode]#012"Timeout"#012Detail: connect failed in tcp_connect()
Hi,
A fairly reliable connection failed last night with a connect failure. There's no network failure at the moment, No changes were made afaik at 4am when it started. I've restarted pulseway and rebooted the PC but the problem persists.
What can I do?
Thanks,
-
Hi, Not really a problem any more so much as a curiosity. I have a generic tablet which ran the app fine on kitkat. I installed a customised lollipop rom and google play says "Your device isn't compatible with this version". I downloaded it via 'amazon underground' and pulseway installed and fired up perfectly.
It seemed possible that my country ID was restricting the download, but it lets me get bbc iplayer so i've discounted that.
As the device is the same, and no other apps seem incompatible, I'm not sure what else to look for.
pulseway on amazon is 4.6.5 rather than 4.8 and I doubt I will have 'amazon underground' running as a service so updating will be manual.
If the google play restriction cannot be identified, perhaps the website could have the most recent apk to download?
Thanks,
-
-
Thanks Chris, that's good to know for future use - but as the 4.6.5 version fails whether it updates or not, is there a way to stop the 4.4 version updating? If not, is there a way to debug the error. And if not that, what do you suggest we do?
-
Thanks... Shame they haven't succeeded yet because all it really needs is a flag to prevent auto-upgrades - essential for any enterprise-level software! - and since they already know much about the system (by definition) it shouldn't be too hard to figure out the common factors of those that won't or can't upgrade.
-
Is any part of that reply going to help me and others?
-
Hi support - do you have an approved way of preventing the upgrade and running the previous, working version?
Thanks,
-
"Do you have any particular software that will change default core libs (glibc, gcc, kernel, OpenSSL) ?"Not that I'm aware of (just an old system):[root@b3 17:29:15] $ ldd --versionldd (Debian EGLIBC 2.11.3-4) 2.11.3[root@b3 17:30:45] $ uname -a; cat /etc/debian_version; openssl versionLinux b3 2.6.32-5-686 #1 SMP Tue May 13 16:33:32 UTC 2014 i686 GNU/Linux6.0.10OpenSSL 0.9.8o 01 Jun 2010
The REST API is working if that eliminates anything (such as data collection rather than data reporting)...
-
Thanks, not sure how I missed your topic!
it seems to have upgraded itself at 17:53 13th Oct on my box
2518926 Oct 13 17:53 /usr/sbin/pulsewayd17162 Oct 13 17:53 /usr/sbin/pulseway-updateIf I copy the old version back in:2199212 Oct 16 13:55 pulsewayd17162 Oct 16 13:55 pulseway-updateIt logs in OK as 4.4.3, and then upgrades itself to 4.4.52518926 Oct 16 13:59 pulsewayd17162 Oct 16 13:59 pulseway-updateAnd crashes:Oct 16 13:59:23 b3 pulseway: Configuration read successfully!Oct 16 13:59:31 b3 pulseway: Got exception signal: 4One problem is, of course, that signal:4 could be anything so there's no way of going further without guessingThe other problem is that there's no way of stopping or backing out of the upgrade, which is bound to cause problems in real world usage.At the very least a system manager needs notice that it's going to stop working... -
Thanks Chris, same problem:
[root@b3 09:16:58] $ uname -aLinux b3 2.6.32-5-686 #1 SMP Tue May 13 16:33:32 UTC 2014 i686 GNU/LinuxOct 16 09:16:48 b3 pulseway: Configuration read successfully!Oct 16 09:16:55 b3 pulseway: Got exception signal: 4And a diff of config.xml with account details removed:diff -awB config.xml config.xml.sample9c10,11< <Account Username="myaccount" Password="hex" UseCustomServer="false" CustomServerAddress=""/>---> <Account Username="" Password="" UseCustomServer="false" CustomServerAddress="" />>13c15,16< <ComputerInformation Name="b3new" Group="Default"/>---> <ComputerInformation Name="" Group="Default" />>138d165< </config>139a167> </config>I then change all true to false:< <OnShutdown Priority="0" Enabled="false"/>---> <OnShutdown Priority="0" Enabled="true" />< <SSLCertificateExpiration Days="15" Priority="0" Enabled="false"/>---> <SSLCertificateExpiration Days="15" Priority="0" Enabled="true"/>< <WebSiteNotAvailable Minutes="1" Priority="0" Enabled="false"/>---> <WebSiteNotAvailable Minutes="1" Priority="0" Enabled="true"/>>And same problemOct 16 09:25:55 b3 pulseway: Configuration read successfully!Oct 16 09:26:02 b3 pulseway: Got exception signal: 4Can you give me an idea of what signal 4 means? -
Thanks Mark,uname -a; cat /etc/debian_version; openssl versionLinux b3 2.6.26-2-686 #1 SMP Wed Sep 21 04:35:47 UTC 2011 i686 GNU/Linux6.0.6OpenSSL 0.9.8o 01 Jun 2010For the config.xml I've copied the sample, typed in the account and plaintext password, and put a ComputerInformationName in - left everything else. After failing, a diff between config.xml and config.xml.sample shows the plaintext password changed and the xml 'tidied up'.I can remove/reinstall pulseway over ssh but is there a way of diagnosing the fault beforehand?Thanks,
-
Hi,
It seemed to be working until the 13th Oct (was it auto-upgraded? - the config.xml.sample has the same datestamp as the first failure. the 'my account' 'monitored systems' shows it last logged in as 4.5.3 )
Now I can't get beyond:
pulseway: Configuration read successfully!pulseway: Got exception signal: 4Originally this happened with the existing config.xml. I copied the config.xml.sample, added in my Account details and gradually changed "true"s to "false"sI can connect using the rest apiI can't seem to find any clue as to what signal 4 is? Is there a debug option?Thanks, -
Perfect... thanks.
Re. the title of the notification - is that meant to appear on the dashboard notification?
-
An offline notification will be sent, please set the notify_when_offline to true as specified here:
http://www.pulseway.com/restapi/#publish
Once the instance goes offline it could take up to 15 minutes for the notification to be fired.
Marius
Thanks but I can't see any mention of that - just this:
"next_refresh_interval_minutes": 5
-
A couple more queries...
If I send a notification I can't see the title in the dashboard/mobile (but it does show on the email subject). On the dashboard I see the date/time and no reference to the title text at all.
If I send HTML tags they are ignored (or rather the tags are shown as text) in the description of the dashboard/mobile, but are interpreted as tags in my gmail - I'm guessing this is gmail's doing - but having some sort of text highlighting would be useful - is there any I can use?
-
Thanks Paul - yes, I didn't notice that but it works fine now!
Do you know about notifications if the refresh-interval expires?
-
Hi, I'm using the Publish API to create a system, which shows up on the dashboard. I then send a notification using curl:
https://api.pulseway.com/v1/publish -H Content-Type:application/json -u <user;key> -X POST -d '{"instance_id":"v7black","title":"Test error","message":"Running out of disk space","priority":"critical"}'which gets a valid response: {"response_code":200,"error_message":null}But I can't see any notification on the dashboard/mobileThe name/description/group I used in publish seems to have been lostI put the publish again, and the group/description is back... however once the 'refresh interval' expires it greys out but there is no notification.thanks for any pointers... -
Thanks Marius
-
Hi, do you have a timetable for the REST API? As it stands I am interested in doing a 'proof-of-concept' but can imagine several things I'd like to see before it became useful as a central monitoring point.
Off the top of my head...
1)Notifications that prompt for a reply
eg 'Loss of power for 5 minutes and and UPS down to 20 minutes. Power off non-essential subsystems? (Y/N)') - that can be replied to over a mobile .
The difference between this and 'send a command' would be application-specific rather than system-specific, for instance in the middle of a backup you might get a notification like 'Expected volume MONDAY but volume TUESDAY is loaded. 'C'ancel backup or 'O'vewrite data on TUESDAY'
2) A drilldown, eg:
Storage --> Mounts --> Disk units
for example - Storage goes to a list of local mounts/volumes, and each of them shows lower-level disks which collectively make up the volume, finally ending in the lowest-level disk 'properties' which are useful.
Ideally though, the drilldown will be the presentation layer, not the data-layer - ie the data item could be supplied in any order, and updated in any order but an alternative way to drill down could be set rather than fixed in the API.
for instance a group of systems can have a 'Storage' icon, which leads to a screen listing disk usage for each 'root partition' of each system within the group - allowing you to see multiple systems at a glance.
Thanks
connect failed in tcp_connect()
in Linux
Posted
fixed - ws0.pulseway.com was inaccessible via just my ISP for 8 hours, but fine using any other ISP.