Usher server down: "Very early error"

[corneliusroemer]

I was just trying to use Usher and got the following server error (both for usher.bio and dev.usher.bio [the test server]):

Couldn't connect to database hgcentral on hgnfs1-c as hgcentuser. Can't connect to MySQL server on 'hgnfs1-c' (115) We hate this error more than any other and may be already looking into it. You can help us by telling us about it, the email is [email protected]. We will fix it ASAP. And even if this server is failing right now, usually, one of our other three international mirrors is still working. The three mirrors are genome.ucsc.edu (US), genome-euro.ucsc.edu (Germany), genome-asia.ucsc.edu (Japan). You may not find your custom tracks and saved sessions there, but using another mirror should allow continuing your work while we are fixing the problem.

[corneliusroemer]

Oh nice, I didn't know about the Euro mirror, that might be a better option for me anyways!

https://genome-euro.ucsc.edu/cgi-bin/hgPhyloPlace

This seems to work!

[corneliusroemer]

Hmm ok it fails soon after submitting sequences:

[maximilianh]

Bad weather today, and power is out on the entire UCSC campus, the servers are on generators, but there is something still off. It'll probably take a few hours to look into the problem and fix it. In the meantime, you can use https://genome-euro.ucsc.edu. Sorry for the problem.

[maximilianh]

Oh sorry I read too quickly. If the Euro mirror doesn't work, then I have no alternative right now. We'll have to wait for a few hours until someone on campus can look into this, sorry.

[russcd]

I was able to get the expected output for a few sequences from the usher.bio. I wonder if this is something specific to the sample(s) you're looking at? Can you share the file you're trying to upload?

…

On Mon, Feb 5, 2024 at 5:44 AM Maximilian Haeussler < ***@***.***> wrote: Oh sorry I read too quickly. If the Euro mirror doesn't work, then I have no alternative right now. We'll have to wait for a few hours until someone on campus can look into this, sorry. — Reply to this email directly, view it on GitHub <#363 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACMZAMKDTBSYBA6HGJVUQCDYSDO4FAVCNFSM6AAAAABC2ETVSKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMRXGA2DKMJRGU> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- Russ Corbett-Detig Associate Professor Department of Biomolecular Engineering University of California, Santa Cruz

[maximilianh]

The server is back up now, the admins fixed it half an hour ago. On Mon, Feb 5, 2024 at 4:36 PM Russell Corbett-Detig < ***@***.***> wrote:

…

I was able to get the expected output for a few sequences from the usher.bio. I wonder if this is something specific to the sample(s) you're looking at? Can you share the file you're trying to upload? On Mon, Feb 5, 2024 at 5:44 AM Maximilian Haeussler < ***@***.***> wrote: > Oh sorry I read too quickly. If the Euro mirror doesn't work, then I have > no alternative right now. We'll have to wait for a few hours until someone > on campus can look into this, sorry. > > — > Reply to this email directly, view it on GitHub > <#363 (comment)>, or > unsubscribe > < https://github.com/notifications/unsubscribe-auth/ACMZAMKDTBSYBA6HGJVUQCDYSDO4FAVCNFSM6AAAAABC2ETVSKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMRXGA2DKMJRGU> > . > You are receiving this because you are subscribed to this thread.Message > ID: ***@***.***> > -- Russ Corbett-Detig Associate Professor Department of Biomolecular Engineering University of California, Santa Cruz — Reply to this email directly, view it on GitHub <#363 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AACL4TJ2A3TD6Y6L72KDW4LYSD4A3AVCNFSM6AAAAABC2ETVSKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMRXGI3TMMJXGI> . You are receiving this because you commented.Message ID: ***@***.***>