# Problem with upgrade to 6.2.15 from 6.1.x version

**URL:** <https://forums.foundationdb.org/t/problem-with-upgrade-to-6-2-15-from-6-1-x-version/2017>\
**Category:** Using FoundationDB\
**Created:** [March 21, 2020, 1:35am UTC](https://forums.foundationdb.org/t/problem-with-upgrade-to-6-2-15-from-6-1-x-version/2017 "2020-03-21T01:35:42Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![kentwang929](https://sea1.discourse-cdn.com/foundationdb/user_avatar/forums.foundationdb.org/kentwang929/32/799_2.png) [@kentwang929](https://forums.foundationdb.org/u/kentwang929)\
**Post date:** [March 21, 2020, 1:35am UTC](https://forums.foundationdb.org/t/problem-with-upgrade-to-6-2-15-from-6-1-x-version/2017/1 "2020-03-21T01:35:42Z")

</div>

As the title suggests, after installing the 6.2.15.pkg, fdbcli is now complaining

`Could not communicate with a quorum of coordination servers: 127.0.0.1:4689 (unreachable) `  
I noticed that on activity monitor that I do not see fdbserver running, only 2 fdbmonitor processes. I have tried restarting the fdbserver with launchctl multiple times, but to no avail. I’ve also checked with lsof to see if 4689 is occupied, and don’t find any entry.

What else could be the issue?

---

<div class="post-metadata">

**Author:** ![alexmiller](https://sea1.discourse-cdn.com/foundationdb/user_avatar/forums.foundationdb.org/alexmiller/32/326_2.png) [@alexmiller](https://forums.foundationdb.org/u/alexmiller)\
**Post date:** [March 21, 2020, 1:58am UTC](https://forums.foundationdb.org/t/problem-with-upgrade-to-6-2-15-from-6-1-x-version/2017/2 "2020-03-21T01:58:51Z")

</div>

Oh, well, apparently fdbmonitor isn’t piped or saved anywhere, which makes this a little more troublesome.

In `/usr/local/foundationdb/logs/` there are log files. Does the most recent one happen to have any `Severity="40"` lines in it?

Assuming you haven’t changed anything, fdbmonitor should be equivalent to manually running:  
`sudo /usr/local/libexec/fdbserver --public_address=auto:4689 --listen_address=public --datadir=/usr/local/foundationdb/data/4689 --logdir=/usr/local/foundationdb/logs`

Could you try doing that locally to see what the error is?

---

<div class="post-metadata">

**Author:** ![kentwang929](https://sea1.discourse-cdn.com/foundationdb/user_avatar/forums.foundationdb.org/kentwang929/32/799_2.png) [@kentwang929](https://forums.foundationdb.org/u/kentwang929)\
**Post date:** [March 21, 2020, 7:57pm UTC](https://forums.foundationdb.org/t/problem-with-upgrade-to-6-2-15-from-6-1-x-version/2017/3 "2020-03-21T19:57:22Z")

</div>

Did a `cat trace.127.0.0.1.* | grep "Severity=\"40"` and did not get anything.

Running  
`sudo /usr/local/libexec/fdbserver --public_address=auto:4689 --listen_address=public --datadir=/usr/local/foundationdb/data/4689 --logdir=/usr/local/foundationdb/logs`  
starts the server normally. No error. Cluster quickly becomes healthy after a quick reinitialization.

I guess something is borked with start up?

`sudo launchctl load /Library/LaunchDaemons/com.foundationdb.fdbmonitor.plist` is what i used to start the service following [https://apple.github.io/foundationdb/administration.html](https://apple.github.io/foundationdb/administration.html)

---

<div class="post-metadata">

**Author:** ![kentwang929](https://sea1.discourse-cdn.com/foundationdb/user_avatar/forums.foundationdb.org/kentwang929/32/799_2.png) [@kentwang929](https://forums.foundationdb.org/u/kentwang929)\
**Post date:** [March 30, 2020, 10:22pm UTC](https://forums.foundationdb.org/t/problem-with-upgrade-to-6-2-15-from-6-1-x-version/2017/4 "2020-03-30T22:22:19Z")

</div>

So turns out, launchctl is not happy when there are foundationsdb.conf and foundationsdb.conf.bak in /usr/local/etc/foundationdb directory. After removing one of them and restarts fdb, the server is now launched successfully. In activity monitor, there is now one fdbmonitor and one fdbserver process as opposed to two fdbmonitor processes before. This seems like a bug to me, but nevertheless, the issue is resolved now.