Thank so much Tim, At no time during the 6.x versions does it tell you to run "database migrate ignored". All of the instructions ignore the fact that someone that started with version 6.x will not have had those migrations run, and yet are forced to do so to upgrade to 7.x, where they will fail.This gives me hope though that skipping them may yield a completely workable 7.x system. So I'm going to try to skip the other 5.x migration that's now holding us up since we skipped the other. (we modified the 5.7 DB script called by the cocoon files, that creates the un-needed index, to basically have a null output)
I can also try upgrading through the 6.x versions, but I see no reason this will succeed, only a slower path to get us to where we are (it failed with 6.2 and 6.3, I only haven't tried 6.1). Any other suggestions form today? I apologize for taking up so much of your time. I have a schedule of 'upgrade DB to 7' for end of October, so I am actively pursuing a solution. Another idea was to perform some sort of WP-style export of the data, then re-import to individual table on a clean 7 install. Dspace does not have that many data tables, so it's possible, but unwieldy and very far from ideal, or maybe even no possible. On Friday, October 15, 2021 at 9:39:21 AM UTC-5 Tim Donohue wrote: > Hi Steven, > > I apologize, but I haven't had time to get back to this yet. It is > definitely an odd error to be occurring, as I would have expected this > migration to have run *before* you attempted an upgrade to 7.0. > > One option might be to go back to your 6.x site and run the "dspace > database migrate ignored" script there (using the 6.x codebase). If it > works there, it might be enough to get your database *past* that > migration (once it is run, it never runs again). Then you could attempt > the upgrade to 7.0. > > You may be correct that there's a migration bug at play here... I honestly > haven't had time to see if I can replicate this elsewhere. The expectation > though is that *ideally* all pre-7.0 migrations should have already been > run on your database (even those which are returned as "ignored")....as > many upgrades involve running that "dspace database migrate ignored" > command (which should have previously triggered those ignored migrations to > run). > > The * oddity* here is that somehow you (and a few others on this list) > have an older 5.7 migration that wasn't triggered until the 7.0 > upgrade...that's unexpected to me as well. Generally, we expect 5.x > migrations to trigger during 5.x minor upgrades or, at worst, during an > upgrade from 5.x to 6.x. The fact that it was never triggered before is > very odd, as it's obviously no longer a valid migration as you move to 7.x > -- it only worked on a 5.x style database. > > I'll see if I can get back to this sometime next week...apologies though, > as this comes during a very busy couple of weeks. I'll also see if I can > ask around to find another developer with time to debug what could be going > on here. > > Thanks again for reporting this...hopefully we can find a way to > workaround this soon. > > Tim > ------------------------------ > *From:* [email protected] <[email protected]> on behalf > of S. T. <[email protected]> > *Sent:* Thursday, October 14, 2021 3:09 PM > > *To:* DSpace Technical Support <[email protected]> > *Subject:* Re: [dspace-tech] Re: Seemingly crazy catch 22 Dspace DB > migration loop question!!! > > Hi Tim - did you see the returned DB info data? Do you have any > suggestions? We are running into a brick wall over here, and so far, we are > unable to resolve any issues brought about by upgrading using the 'ignored' > flag. We are not only stopped at this issue, but also at other multiple > issues involving DB modification using an index. I commented out the 5.7 db > fix, and the script ran fine, but failed on another, similar issue. > > So, to me, this says that we have an unexpected or non-standard table > structure? There is no reason why our table structure/indexes should be > non-standard and unexpected, we have done nothing custom to the database. > > Any suggestions? > > On Thursday, October 14, 2021 at 11:04:30 AM UTC-5 Tim Donohue wrote: > > Hi Steven, > > What does your "dspace database info" command return? > We'd need more information here to help debug it. This seems like a very > odd error to me as well, as that migration was initially added in 5.7 and > ported to 6.1... so, if you were already running 6.1 or above, this > migration *should have already run for you*. > > That said, it's possible your "dspace database info" command would provide > us with additional hints. I'm assuming that somehow this migration may > not have ever been applied to your 6.x site....and now that you are > updating to 7.x, it's causing issues. > > Tim > ------------------------------ > *From:* [email protected] <[email protected]> on behalf > of S. T. <[email protected]> > *Sent:* Thursday, October 14, 2021 10:47 AM > *To:* DSpace Technical Support <[email protected]> > *Subject:* [dspace-tech] Re: Seemingly crazy catch 22 Dspace DB migration > loop question!!! > > Thanks Joel, I appreciate the help. I guess I should search the Jira for > Dspace on their github repo, and / or submit this bug. I am assuming they > have a Jira and a github repo.Let me run those commands, and I will share. > > On Thursday, October 14, 2021 at 7:56:24 AM UTC-5 [email protected] wrote: > > Steven, > > I claim to be no expert, but maybe you (and another post > <https://groups.google.com/g/dspace-tech/c/1paSeuUkjFo> from September) > stumbled upon a bug? Can you share the results of the following commands? > > > *[dspace]/bin/dspace database validate * > *[dspace]/bin/dspace database info* > > It's suspicious that you started on version 6 and the Flyaway migration is > attempting to issue a patch for something that existed in version 5.7. > > Just for fun, you might want to try this, too, if you haven't already. > > > *[dspace]/bin/dspace database repair * > > --Joel > > > > > On Wednesday, October 13, 2021 at 5:47:46 PM UTC-4 [email protected] > wrote: > > We are trying to upgrade our system from 6 to 7, we are following all the > steps. > > We get this message on DB migration: > Migration > V5.7_2017.04.11__DS-3563_Index_metadatavalue_resource_type_id_column.sql > failed > > When we attempt to update the DB, if fails right there, because it is (?) > trying to create a field WHICH APPARENTLY DOES NOT EXIST in 6.x or 7.x from > a previous field THAT NEVER EXISTED ON OUR 6.x install. > > This script, which is part of the Flyway DB upgrade auto process > > > org/dspace/storage/rdbms/sqlmigration/postgres/V5.7_2017.04.11__DS-3563_Index_metadatavalue_resource_type_id_column.sql > > Contents are: > DROP INDEX IF EXISTS metadatavalue_resource_type_id_idx; CREATE INDEX > metadatavalue_resource_type_id_idx ON metadatavalue (resource_type_id); > > However, WE started with V 6.0. So, (1) that column did not exist in the > 6.0 schema. (2) Furthermore, the field it is trying to create USING the > value from that 5.7 field, also DOES NOT EXIST in the schema for 6.xx OR > 7.xx. > > So, the migration script seems to be trying to alter the metadatavalue > table using a column which doesn't exist to create a column that doesn't > apparently exist in anyone's backend. > > Can anyone who either knows migrations very well, or the Dspace backend > care to comment or assist? This is so super frustrating and confusing. What > are we missing? > > Thanks so much > > Steven Turner > > -- > All messages to this mailing list should adhere to the Code of Conduct: > https://www.lyrasis.org/about/Pages/Code-of-Conduct.aspx > --- > You received this message because you are subscribed to the Google Groups > "DSpace Technical Support" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/dspace-tech/6fe30d91-4b79-48c6-aa30-2e13ce040132n%40googlegroups.com > > <https://groups.google.com/d/msgid/dspace-tech/6fe30d91-4b79-48c6-aa30-2e13ce040132n%40googlegroups.com?utm_medium=email&utm_source=footer> > . > > -- > All messages to this mailing list should adhere to the Code of Conduct: > https://www.lyrasis.org/about/Pages/Code-of-Conduct.aspx > --- > You received this message because you are subscribed to the Google Groups > "DSpace Technical Support" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/dspace-tech/4a8d18a7-1948-4c39-bfab-bcf44b3a1957n%40googlegroups.com > > <https://groups.google.com/d/msgid/dspace-tech/4a8d18a7-1948-4c39-bfab-bcf44b3a1957n%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- All messages to this mailing list should adhere to the Code of Conduct: https://www.lyrasis.org/about/Pages/Code-of-Conduct.aspx --- You received this message because you are subscribed to the Google Groups "DSpace Technical Support" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/dspace-tech/22a71cda-9dd0-450f-a4e7-5826b89d0e3bn%40googlegroups.com.
