This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
Leonard, > How does [removal of data-products from the product-queue] work? I'm not > seeing it in the reference pages. Removal of data-products from the product-queue isn't mentioned because it's automatic and not under the user's control. > The file > I previously inserted is still in there, days later. Data-products are automatically removed on a least-recently-used basis in order to make room for new data. Apparently, you haven't added much additional data. > How does one set > the expiration up? One doesn't. > Here's a product going in, on mlsoserver (upstream): > > $ ldmadmin watch > (Type ^D when finished) > Jun 05 20:17:26 pqutil INFO: 103020480 20120605201726.058 EXP 000 > 20120603.145219.FTS > > Here it is on mlsodata (downstream): > > $ ldmadmin watch > (Type ^D when finished) > Jun 05 20:18:28 pqutil INFO: 103020480 20120605201726.058 EXP 000 > 20120603.145219.FTS > > Then, here is where it should be: > > $ egrep -v '^#' pqact.conf > EXP ^(.*) FILE -overwrite -flush -close > -log /export/data1/Data/ldm/\1 > > !$ > ls -l /export/data1/Data/ldm > total 100608 > -rw-rw-r-- 1 ldm6 cordyn 103020480 Jun 5 14:18 20120603.145219.FTS > > So, it's working now! Great! > I don't know what changed. I simply started it up today and it works. Ah, computers! :-) > The file from the last test is still in there, and the one I just > inserted is still in there, too. I would like them to be removed after > being sent, I guess. The only way to remove them would be to add more data-products until they're removed in order to make room. > upstream: > $ pqmon > Jun 05 20:25:10 pqmon NOTE: Starting Up (3587) > Jun 05 20:25:10 pqmon NOTE: nprods nfree nempty nbytes maxprods > maxfree minempty maxext age > Jun 05 20:25:10 pqmon NOTE: 2 1 122067 103020672 4 > 2 122065 396982144 514464 > Jun 05 20:25:10 pqmon NOTE: Exiting Only 2 data-products. > downstream: > $ pqmon > Jun 05 20:26:39 pqmon NOTE: Starting Up (19920) > Jun 05 20:26:39 pqmon NOTE: nprods nfree nempty nbytes maxprods > maxfree minempty maxext age > Jun 05 20:26:39 pqmon NOTE: 1 1 122068 103020600 4 > 2 122065 396982216 491 > Jun 05 20:26:39 pqmon NOTE: Exiting Only 1 data-product. > I know the nprods is the number of *slots*, Actually, "nprods" from pqmon(1) is the number of data-products in the product-queue. > but pqcat still gets the > files downstream, so they are still "in there", if that is meaningful: > > $ pqcat > bozo > Jun 05 20:28:06 pqcat NOTE: Starting Up (19941) > Jun 05 20:28:06 pqcat NOTE: Exiting > Jun 05 20:28:06 pqcat NOTE: Number of products 1 > [ldm6@mlsodata ~]$ ls -l bozo > -rw-rw-r-- 1 ldm6 ldm 103020480 Jun 5 14:28 bozo Looks good to me. > There was one issue from the log that I fixed before inserting the file > to find that LDM is now working: > > Jun 5 12:52:22 mlsodata pqact[19098] ERROR: cannot chdir > to /export/data1/Data/ldm-tmp/data: No such file or directory > > That's the downstream system. Note that the directory in the log > message is not the directory in the pq configuration, which is > /export/data1/Data/ldm That's a fairly major problem. Glad you got it fixed. > Also, both systems complain at startup that > Jun 5 12:57:59 mlsodata pqact[19384] NOTE: Starting Up > Jun 5 12:57:59 mlsodata pqact[19384] WARN: No such file or directory > Jun 5 12:57:59 mlsodata pqact[19384] WARN: stateRead(): Couldn't open > "/home/ldm/etc/pqact.conf.state" > Jun 5 12:57:59 mlsodata pqact[19384] WARN: Previous-state information > doesn't exist. Continuing... > > This is because the account name is ldm6 (we're upgrading while the > current system is using the account ldm). The above messages are because there are no .*.state files in the etc/ subdirectory corresponding to the pqact(1) configuration-files. Consequently, the pqact(1) processes don't know where they left-off at the end of the previous session. > -- > ==Leonard E. Sitongia > High Altitude Observatory > National Center for Atmospheric Research > P.O. Box 3000 Boulder CO 80307 USA > address@hidden voice: (303)497-2454 fax: (303)497-1589 Regards, Steve Emmerson Ticket Details =================== Ticket ID: XBC-274988 Department: Support LDM Priority: Normal Status: Closed