WS-26-500-DC Rebooting

[Kingsley]

Has the WS-26-500-DC got the same hardware fault as the small percentage of the WS-6 and WS-8 models which were all possibly affected by 3.3V CAP issue in a previous batch?


I have a site that had a WS-12-250-DC for 6+ months run perfectly fine. Switch was upgraded recently for more ports to the 26 port switch on 1.4.7 and now every few days the switch reboots. Does exactly as explained in another post. Switch randomly reboots and log shows a normal startup once I can log back in. The site has external battery monitoring and a massive battery bank at 12v so it is NOT a power issue.
On the other hand I have another WS-26-500-DC on a different site that is running solid and running a whole lot more equipment than the switch that is rebooting. This first switch was bought around the start of the year how ever this second switch that is rebooting is only a couple months old.
Any ideas? Can I roll the firmware back to something else?

[Dave]

The WS-26-DC models have no known hardware issues.

How are you charging your battery's?

Can you post up your power tab also, the one that has the hibernate voltage settings.

[sirhc]

You should be running v1.4.8rc11 or the latest code as there are IMPORTANT fixes for DC switches - read release notes.
You're saying your powering the WS-26 with 12V DC?
PLEASE upgrade your DC switches to v1.4.8rcX, in fact unless you want to run known bugs I would upgrade all switches to the latest version.
"RC" does not mean release candidate to us.
The day we released v1.4.7 as soon as someone reported a bug a fix was made and v1.4.8rc1 was born.
v1.4.8rcX has all the known bugs reported and fixed since v1.4.7 was released.
After upgrading your switches please detail the power source:
How many batteries
How are they interconnected (series or parallel)
What is charging them
Does the switch connect DIRECTLY to the battery bank or is there something in between?

12V is NOT the best/efficient input voltage to use, we recommend 24V and up, the fact that is runs down to 9V is to allow deeper discharge in an emergency.
Post up screen grabs of the following TABs:

Status
Ports
Power
Device/Status

[Kingsley]

During the day the site is charged by solar. There is something like 12 or 1400 Ah of batteries connected in parallel at 12v and there is a generator that runs 12 hours over night every night. The switch runs from a 12v power distribution box which is connected directly to the batteries. It is not a power issues, there is a lot of equipment running out of the same distribution box that does not see this issue. Even the previous switch the WS-12-250-DC did not see this issue running the same firmware. I don't believe we have added any extra POE devices since upgrading the switch, it has only had non POE devices added.

I agree that it would be better at 24v how ever there is equipment on this site that requires 12v. Yes I could step up the voltage but your switch can handle 9-72v I shouldn't need to run extra voltage equipment. As I keep saying the 12 port switch was perfectly fine. I have management of 50+ netonix switches on both 12v and 24v systems yet this is the only switch I am having problems with. I believe it all points to this particular switch. Now I do not know if it is hardware or software issue hence I am posting here.

My boss is always doubtful upgrading firmware to anything other than listed as a stable release. How ever if you are confident the latest RC firmware is stable enough I can look at running it. This switch is linking DMR radio repeaters together so it is paramount there is zero issues as it provides communication and emergency radio comms into areas there is no cell phone coverage.

Please see attached screen shots as requested.

[sirhc]

We have discussed this MANY times.

So all the bugs reported since the release of v1.4.7 have been fixed in v1.4.8rc11 so why would you not want to use the firmware that has all the bugs fixed?


viewtopic.php?f=17&t=3136&p=21016&hilit=+bugs+reported+fixed#p21016
viewtopic.php?f=17&t=2745&p=20986&hilit=+bugs+reported+fixed#p20986
viewtopic.php?f=17&t=2931&p=19985&hilit=+bugs+reported+fixed#p19985


"RC" DOES NOT MEAN RELEASE CANDIDATE AND AFTER V1.4.8 IS FINALIZED WE WILL NO LONGER USE "RC" AS PEOPLE THINK IT MEANS RELEASE CANDIDATE. SO AFTER V1.4.8 THERE WILL BE V1.4.8.X

More importantly please look at release notes in v1.4.8rcX there was an IMPORTANT FIX FOR ALL SMART DC SWITCHES
See below in BIG RED
If the power supply controller chip gets too hot it will shut down, in the FIX it kicks the fans on to prevent this.

NOTE: MSTP is still being tested so use MSTP at your own risk.

FIXED/CHANGED
- Fixed SNMP support for Q-SWITCH-MIB, LLDP and CDP - RC7
- Changed temp sensor readings on DC models to effect fan speeds - RC2
- Grey out unavailable bulk poe options - RC3
- Upper out of voltage range for IDC is now 54V - RC4
- Fix for IGMP snooping with multiple VLANs - RC9
- Fixed DC calibration restarting POE - RC10
- Fixed problem with DHCP snooping affecting fragmented UDP packets - RC10
- Fixed problem with ERPS enabling forwarding a failed port - RC10
- Fixed problem with MSTP setting priority and path cost - RC10
- Fixed SSH login with RADIUS when the password contains special characters - RC11
- Fixed scheduled port bounce - RC11


ENHANCEMENTS
- Added ERPS protocol implementation is g.8032v2 - Ring Protection - RC1
- Added BETA note to ERPS - RC3
- Added port name to "show interface status" CLI command - RC3
- ADD an input voltage calibration for DC models - RC4
- Improved Q-BRIDGE-MIB support - RC5
- Improved fan control on DC models - RC6
- Improved stability of ERPS- RC6
- Added "Enable Power" action to watchdog - RC8
- Added ifMtu to IF-MIB - RC10
- Added dot3StatsDuplexStatus from EtherLike-MIB - RC10
- Added the time counters were cleared to port detail - RC10
- Added watchdog description to email notification - RC10
- Added terminal length option to CLI - RC10
- Added ability to re-order VLANs using drag and drop - RC10
- Enabled IGMP proxy for IGMP snooping - RC11
- Added "show igmp" CLI command - RC11


KNOWN ISSUES
- Some language templates need help - please private message Eric Stern to help

I know you say the previous switch used the same "transfer" box but the 500 is a bigger power supply with a much bigger possible draw and inrush than the 250. We designed the switches to hook directly to the battery bank with nothing in between except maybe a fuse.

Things I would try in this order:
Upgrade the firmware
Hook switch DIRECTLY to battery bank
Swap switch out with identical unit (spare) and if problem is still there then it is not the switch.

[Kingsley]

So what is RC standing for in current firmware releases?

Upgraded firmware to 1.4.8rc11 and it continues to reboot. Time to RMA this unit and jump in a helicopter to replace it with a new one.

[sirhc]

So what is RC standing for in current firmware releases?

Upgraded firmware to 1.4.8rc11 and it continues to reboot. Time to RMA this unit and jump in a helicopter to replace it with a new one.

Way back in the beginning it meant Release Candidate but that changed years ago since we try to fix bugs ASAP.

After v1.4.8 if finalized we will stop the naming scheme we use and simply uses dots so in the future it will v1.4.8.1 and so on

[sirhc]

Well if you replace it with another unit and the problem persists we will need to look a something else as the cause?