- users-list - monetdb.org

Announcement: New Jan2022-SP2 Bugfix release of MonetDB suite
by Sjoerd Mullender 08 Apr '22

08 Apr '22

The MonetDB team at MonetDB Solutions BV is pleased to announce the Jan2022-SP2 bugfix release of the MonetDB suite of programs. More information about MonetDB can be found on our website at <https://www.monetdb.org/>. For details on this release, please see the release notes at <https://www.monetdb.org/documentation/admin-guide/release-notes/jan2022/>. As usual, the download location is <https://www.monetdb.org/downloads/>. ### Client Package - Improved the handling of the \r (internal pager) command in mclient. It now properly counts the header of table, and when a (very) long table is being printed and aborted part way in the built-in pager, not all data is transferred to the client (and then discarded). Instead at most 1000 rows are transferred. ### MonetDB Common - Improved speed of BATappend to empty varsized bat: we now just copy the heaps instead of inserting individual values. - Improved speed of projection (BATproject) on varsized bats by sharing the data heap (vheap). - Fixed a race condition which could cause a too large size being written for a .theap file to the BBP.dir file after the correct size file had been saved to disk. - We now ignore the size and capacity columns in the BBP.dir file. These values are essential during run time, but not useful in the on-disk image of the database. - Fixed a bug in the append code for msk (bit mask) bats. - Conversions from floating point types to integral types that involve multiplication now use the "long double" as intermediate type, thereby loosing as few significant bits as is feasible. - Found and fixed another source for the now infamous BBPcheckbats error that sometimes occurs at startup of the server. ### MonetDB5 Server - Improved parsing speed of blob values, especially on Windows. On Windows, using the locale-aware functions isdigit and isxdigit is comparatively very slow, so we avoid them. ### Bug Fixes - [7252](https://github.com/MonetDB/MonetDB/issues/7252): Segmentation fault on second run - [7253](https://github.com/MonetDB/MonetDB/issues/7253): Extremely slow INSERT INTO <table> SELECT - [7254](https://github.com/MonetDB/MonetDB/issues/7254): Commit with deletions is very slow - [7263](https://github.com/MonetDB/MonetDB/issues/7263): PRIMARY KEY constraint is not persistent through server restarts - [7267](https://github.com/MonetDB/MonetDB/issues/7267): Update after delete does not update some rows - [7278](https://github.com/MonetDB/MonetDB/issues/7278): BUG when there is more than one field/filter in the having clause

1 0

wrong command on deb package installation page
by Manuel Pagliai 04 Mar '22

04 Mar '22

Hello, I was following these instructions: https://www.monetdb.org/downloads/deb/ I think there is a mistake in the page and the command sudo wget --output-document=/etc/apt/trusted.gpg.d/monetdb.gpg /downloads/MonetDB-GPG-KEY.gpg should be replaced with sudo wget --output-document=/etc/apt/trusted.gpg.d/monetdb.gpg https://www.monetdb.org/downloads/MonetDB-GPG-KEY.gpg Best regards, *Manuel Pagliai* Visokio Ltd www.visokio.com +44 20 7351 4646

2 1

Iterations in MonetDB
by Oxana Dolmatova 23 Feb '22

23 Feb '22

Hello MonetDB Team, I want to integrate iterations into MonetDB using the barrier symbols provided by MAL (here the variable list_of_updates is repeated body of the iteration and ex_cond is the exit condition): ======================================= res_body = stmt_list(sql->sa, list_of_updates); stmt *s_begin = stmt_begin(sql->sa); list_append(res_list, s_begin); list_append(res_list, ex_cond); stmt *s_check = stmt_check(sql->sa, ex_cond, s_begin); list_append(res_list, s_check); list_append(res_list, res_body); list_append(res_list, stmt_redo(sql->sa, s_check)); list_append(res_list, stmt_exit(sql->sa, s_begin)); return stmt_list(sql->sa, res_list); ======================================== Currently the implementation works, when the exit condition is true, but unfortunately never stops, if the exit condition is false (I assume that this happens due to non-updating computation of ex_cond). Could you please give the reference point in the documentation or code where I could look to check if my assumption is correct? Or point me to a proper mailing-list if this is the wrong one for such questions? :) Thank you and best regards, Oksana

1 0

Fwd: COPY SELECT INTO with problems on new version 11.43.5 (Jan2022)
by Ariel Abadi 16 Feb '22

16 Feb '22

Hi All! I'm resending to the new user list. BR Ariel ---------- Forwarded message --------- From: Ariel Abadi <aabadi(a)starbi.com> Date: Fri, Feb 4, 2022, 18:44 Subject: COPY SELECT INTO with problems on new version 11.43.5 (Jan2022) To: Communication channel for MonetDB users <users-list(a)monetdb.org> Hi All! We found a problem on the new version 11.43.5 (Jan2022). We have some processes that run the COPY SELECT INTO to "move" data from monetdb to other different databases. We are expecting to receive an "answer" from monetdb affected, when the export finishes, like in the example below that informes "3 affected rows" sql>COPY SELECT * FROM cars INTO '/tmp/cars_out.csv' USING DELIMITERS ',' , '\n' , '"';*3 affected rows* While we run the same query on the new server, mclient/monetdb is not returning anything like the image below. sql>COPY SELECT * FROM cars INTO '/tmp/cars_out.csv' USING DELIMITERS ',' , '\n' , '"'; sql> Does anybody have the same issue? Am i doing something wrong? Thks a lot in advance! Ariel

2 1

Announcement: New Jan2022-SP1 Bugfix release of MonetDB suite
by Sjoerd Mullender 14 Feb '22

14 Feb '22

The MonetDB team at MonetDB Solutions BV is pleased to announce the Jan2022-SP1 bugfix release of the MonetDB suite of programs. More information about MonetDB can be found on our website at <https://www.monetdb.org/>. For details on this release, please see the release notes at <https://www.monetdb.org/documentation/admin-guide/release-notes/jan2022/>. As usual, the download location is <https://www.monetdb.org/downloads/>. ### SQL Frontend - [This feature was already released in Jan2022 (11.43), but the ChangeLog was missing] Added SQL procedures sys.vacuum(sname string, tname string, cname string), sys.vacuum(sname string, tname string, cname string, interval int), sys.stop_vacuum(sname string, tname string, cname string). These can be used to vacuum string columns. ### Bug Fixes - [7228](https://github.com/MonetDB/MonetDB/issues/7228): COMMIT: transaction is aborted because of concurrency conflicts, will ROLLBACK instead - [7230](https://github.com/MonetDB/MonetDB/issues/7230): Prepared statement of INSERT with SELECT fails when types difer - [7232](https://github.com/MonetDB/MonetDB/issues/7232): False conflicts when inserting in a not null field - [7237](https://github.com/MonetDB/MonetDB/issues/7237): SELECT with concurrent writes rarely returns corrupt data - [7238](https://github.com/MonetDB/MonetDB/issues/7238): query with system function: "index"(varchar, boolean) fails with GDK error or assertion failure. - [7241](https://github.com/MonetDB/MonetDB/issues/7241): Replacing a view by a query on the view itself crashes the server.

1 0

mailing list software updated
by Sjoerd Mullender 10 Feb '22

10 Feb '22

I have updated the mailing list software on monetdb.org from mailman2 to mailman3. This means that the old accounts for managing list subscriptions do not work anymore. In order to manage your subscription using a web interface, you will need to create a new account at https://www.monetdb.org/lists/. You can associate multiple email addresses to this account and manage all your subscriptions at monetdb.org using a single account, even if they use different email addresses. Unlike in mailman2, in mailman3 the passwords are not saved in plaintext, and therefore they cannot be sent as a monthly reminder. Mailing list archives have been converted to the new system as well. The old web addresses (https://mail.monetdb.org/, https://www.monetdb.org/mailman/ and https://www.monetdb.org/pipermail/) now all get redirected to the new interface. -- Sjoerd Mullender

1 0

None
by None 10 Feb '22

10 Feb '22

keeping the BAT open (I always use "bat("name").XXX", or keep a ref variable in a block. So, as far as I can tell, I should not have 7500 BATs open at once. Am I misunderstanding? Or is there some underlying component/module keeping the BATs open? Or is this just the amount of memory MonetDB will use? I do not have a BAT of BATs (as I was sure that this would cause problems, and from your description, will indeed). Thanks! Ed

1 0

None
by None 10 Feb '22

10 Feb '22

columns in a BAT (ie, BAT[OID, any]), is this just the first BUN? For update operations, does this mean that in the middle of inserting into a set of BATs, that as the BUN count is different, a concurrent query may not realize that the two BATs are synced? (this gets back to multiple client isolation for changes) Thanks for your time! Regards! Ed

1 0

None
by None 10 Feb '22

10 Feb '22

"This is implemented by storing a very large OID for each column. An update to the column destroys this OID."

1 0

None
by None 10 Feb '22

10 Feb '22

"For BAT-sets that contain exactly the same sets of OIDs, those OID-columns can be marked synced(). This information is then used to use positional lookup (the i-th element in BAT1 corresponds with the i-th in BAT2) and to turn semijoin- into very cheap copy-operations, etc." almost makes it sound as if you need synched() to actual test and set this bit. Is this true? How is the "syncedness" ensured when I have a BAT[OID, any], or determined?

1 0