Eugene Klimov issues

Results 220 issues of


                                            Eugene Klimov

clickhouse_backup_last_backup_size_remote show previous value when using `sever --watch`

``` clickhouse@comm-1cbl9-s03-r01:/export/home/dba$ clickhouse-backup list remote full_comm-1cbl9-s03-r01_2025_10_27_15_53_25_UTC 2025-10-27 16:13:20 remote all:34.70GiB,data:44.96GiB,arch:34.70GiB,obj:0B,meta:41.20KiB,rbac:1iB,conf:0B,nc:0B gzip, regular increment_comm-1cbl9-s03-r01_2025_10_28_03_53_25_UTC 2025-10-28 03:53:55 remote +full_comm-1cbl9-s03-r01_2025_10_27_15_53_25_UTC all:736.94MiB,data:25.86GiB,arch:736.92MiB,obj:0B,meta:21.32KiB,rbac7KiB,conf:0B,nc:0B gzip, regular full_comm-1cbl9-s03-r01_2025_10_28_15_53_25_UTC 2025-10-28 16:09:05 remote all:36.36GiB,data:47.02GiB,arch:36.36GiB,obj:0B,meta:61.66KiB,rbac:1iB,conf:0B,nc:0B gzip, regular clickhouse@comm-1cbl9-s03-r01:/export/home/dba$ curl http://127.0.0.1:65080/metrics...

object_disks + --restore-table-mapping can broke already exists table

during downloadObjectDisk part we don't change key names, so when you DROP TABLE , you will corrupt exists table which contains the same object keys

add --skip-empty-tables to restore and restore_remote commands

if the table in the backup is empty, the program would skip restoring it, to avoid wiping out existing data by mistake, the partition filtering feature also should skip the...

properly calculate free space before download for --hardlink-exists-files

After table metadata download, need to calculate required disk space, need check exists files with `checksums` section, need think about `files` section to add `size`

restore engine=Kafka, engine=*Queue, refreshable materialized view, after restore data as `postponedTables`

check uniq types from system.parts_columns shall executed before ALTER TABLE FREEZE for all tables, not individual inside working group to avoid unfreezed data in parallel

moreover, maybe like shadow names shall contains table name and need to execute UNFREEZE in `clean` command

Add options which allow restore non sharded replicated tables

`CLICKHOUSE_SKIP_RESTORE_NON_SHARDED_REPLICATED_TABLES` - false / 0 default value `CLICKHOUSE_SKIP_CHECK_NON_SHARDED_REPLICATED_TABLES_CLUSTER` - with `{cluster}` default value use following query to check global tables ```sql SELECT database, table, if(uniqExact(zookeeper_path) = 1 AND count() >...

replace Fatal to Warning when failed to write to resumable state

need survive during following errors `2025-06-14 02:43:06.649 FTL pkg/resumable/state.go:152 > resumable state: can't write key shard1-20250614000119/shadow/default/hits_raw/default_20250305_145983_165220_26.tar to /var/lib/clickhouse/backup/shard1-20250614000119/upload.state2 error: write /var/lib/clickhouse/backup/shard1-20250614000119/upload.state2: no space left on device`

feature request - github releases with changelog

thanks for your great efforts, nice project could you add to your CI/CD publish to github releases with changelog? it allow "watch"->"releases"

system.tables doesn't work in `clickhouse-server:25.3.6.10034.altinitystable` with DataLake type=glue tables

**Describe the bug** system.tables doesn't work in `clickhouse-server:25.3.6.10034.altinitystable` **To Reproduce** Steps to reproduce the behavior: 1. clickhouse-client -q "SELECT database, name, engine , data_paths , uuid , create_table_query , coalesce(total_bytes,...

antalya